[{"id": "crossner_science_0", "text": "ESA's Advanced Concepts Team also demonstrated theoretically that a deflection of 99942 Apophis could be achieved by sending a spacecraft weighing less than a tonne to impact against the asteroid.", "requested_types": ["COUNTRY", "ASTRONOMICAL_OBJECT", "ORGANISATION"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "ASTRONOMICAL_OBJECT": ["99942 Apophis"], "ORGANISATION": ["ESA", "Advanced Concepts Team"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ASTRONOMICAL_OBJECT, ORGANISATION\n\nText:\nESA's Advanced Concepts Team also demonstrated theoretically that a deflection of 99942 Apophis could be achieved by sending a spacecraft weighing less than a tonne to impact against the asteroid.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ASTRONOMICAL_OBJECT, ORGANISATION\n\nText:\nESA's Advanced Concepts Team also demonstrated theoretically that a deflection of 99942 Apophis could be achieved by sending a spacecraft weighing less than a tonne to impact against the asteroid.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_1", "text": "Studies in human neutrophils first detected a plasma membrane -localized site which reversibly bound 5-oxo-ETE and had the attributes of a Gi alpha subunit -linked G protein-coupled receptor based on the ability of 5-oxo-ETE to activate this class of membrane G proteins by a pertussis toxin -sensitive mechanism.", "requested_types": ["MISC", "CHEMICAL_COMPOUND", "CHEMICAL_ELEMENT", "PROTEIN"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"MISC": ["neutrophils", "plasma membrane -localized site", "pertussis toxin -sensitive mechanism"], "CHEMICAL_COMPOUND": ["5-oxo-ETE"], "CHEMICAL_ELEMENT": [], "PROTEIN": ["Gi alpha subunit", "G protein-coupled receptor", "G proteins"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_COMPOUND, CHEMICAL_ELEMENT, PROTEIN\n\nText:\nStudies in human neutrophils first detected a plasma membrane -localized site which reversibly bound 5-oxo-ETE and had the attributes of a Gi alpha subunit -linked G protein-coupled receptor based on the ability of 5-oxo-ETE to activate this class of membrane G proteins by a pertussis toxin -sensitive mechanism.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_COMPOUND, CHEMICAL_ELEMENT, PROTEIN\n\nText:\nStudies in human neutrophils first detected a plasma membrane -localized site which reversibly bound 5-oxo-ETE and had the attributes of a Gi alpha subunit -linked G protein-coupled receptor based on the ability of 5-oxo-ETE to activate this class of membrane G proteins by a pertussis toxin -sensitive mechanism.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_2", "text": "That first evening session was organized by Jack Yardley from Johns Hopkins University, and included Henry Appelman (University of Michigan), Harvey Goldman (Beth Israel Deaconess Medical Center and Harvard Medical School), Bill Hawk (The Cleveland Clinic), Tom Kent (University of Iowa), Si-Chun Ming (Temple University), Tom Norris (University of Washington), and Robert Riddell (University of Chicago).", "requested_types": ["UNIVERSITY", "PROTEIN", "ORGANISATION", "PERSON", "SCIENTIST"], "distractors": ["PROTEIN"], "gold": {"UNIVERSITY": ["Johns Hopkins University", "University of Michigan", "Harvard Medical School", "University of Iowa", "Temple University", "University of Washington", "University of Chicago"], "PROTEIN": [], "ORGANISATION": ["Beth Israel Deaconess Medical Center", "Cleveland Clinic"], "PERSON": ["Jack Yardley", "Henry Appelman", "Tom Kent"], "SCIENTIST": ["Harvey Goldman", "Bill Hawk", "Si-Chun Ming", "Tom Norris", "Robert Riddell"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, PROTEIN, ORGANISATION, PERSON, SCIENTIST\n\nText:\nThat first evening session was organized by Jack Yardley from Johns Hopkins University, and included Henry Appelman (University of Michigan), Harvey Goldman (Beth Israel Deaconess Medical Center and Harvard Medical School), Bill Hawk (The Cleveland Clinic), Tom Kent (University of Iowa), Si-Chun Ming (Temple University), Tom Norris (University of Washington), and Robert Riddell (University of Chicago).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, PROTEIN, ORGANISATION, PERSON, SCIENTIST\n\nText:\nThat first evening session was organized by Jack Yardley from Johns Hopkins University, and included Henry Appelman (University of Michigan), Harvey Goldman (Beth Israel Deaconess Medical Center and Harvard Medical School), Bill Hawk (The Cleveland Clinic), Tom Kent (University of Iowa), Si-Chun Ming (Temple University), Tom Norris (University of Washington), and Robert Riddell (University of Chicago).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_3", "text": "Viral TK phosphorylates aciclovir into its monophosphate form, which is subsequently phosphorylated to active aciclovir triphoshate by cellular kinase s, thus selectively inhibiting viral DNA polymerase.", "requested_types": ["CHEMICAL_COMPOUND", "ENZYME", "MISC", "PROTEIN"], "distractors": ["PROTEIN"], "gold": {"CHEMICAL_COMPOUND": ["aciclovir triphoshate"], "ENZYME": ["Viral TK", "cellular kinase", "DNA polymerase"], "MISC": ["aciclovir"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ENZYME, MISC, PROTEIN\n\nText:\nViral TK phosphorylates aciclovir into its monophosphate form, which is subsequently phosphorylated to active aciclovir triphoshate by cellular kinase s, thus selectively inhibiting viral DNA polymerase.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ENZYME, MISC, PROTEIN\n\nText:\nViral TK phosphorylates aciclovir into its monophosphate form, which is subsequently phosphorylated to active aciclovir triphoshate by cellular kinase s, thus selectively inhibiting viral DNA polymerase.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_4", "text": "He also won the Copley Medal in 1907, the Henry Draper Medal in 1916 and the Gold Medal of the Royal Astronomical Society in 1923.", "requested_types": ["LOCATION", "AWARD", "DISCIPLINE"], "distractors": ["DISCIPLINE", "LOCATION"], "gold": {"LOCATION": [], "AWARD": ["Copley Medal", "Henry Draper Medal", "Gold Medal of the Royal Astronomical Society"], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, AWARD, DISCIPLINE\n\nText:\nHe also won the Copley Medal in 1907, the Henry Draper Medal in 1916 and the Gold Medal of the Royal Astronomical Society in 1923.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, AWARD, DISCIPLINE\n\nText:\nHe also won the Copley Medal in 1907, the Henry Draper Medal in 1916 and the Gold Medal of the Royal Astronomical Society in 1923.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_5", "text": "Hypomethylation of these genes correlated with decreased expression of various DNMTs and Methyl-CpG-binding domain binding proteins.", "requested_types": ["PROTEIN", "MISC", "EVENT"], "distractors": ["EVENT"], "gold": {"PROTEIN": ["Methyl-CpG-binding domain binding proteins"], "MISC": ["Hypomethylation", "DNMTs"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, EVENT\n\nText:\nHypomethylation of these genes correlated with decreased expression of various DNMTs and Methyl-CpG-binding domain binding proteins.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, EVENT\n\nText:\nHypomethylation of these genes correlated with decreased expression of various DNMTs and Methyl-CpG-binding domain binding proteins.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_6", "text": "In February 2016, he was one of the four scientists of LIGO Scientific Collaboration / Virgo interferometer collaboration presenting at the press conference for the announcement that the first direct gravitational wave observation had been made in September 2015.", "requested_types": ["ORGANISATION", "LOCATION", "EVENT"], "distractors": ["LOCATION"], "gold": {"ORGANISATION": ["LIGO Scientific Collaboration", "Virgo interferometer"], "LOCATION": [], "EVENT": ["first direct gravitational wave observation"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, LOCATION, EVENT\n\nText:\nIn February 2016, he was one of the four scientists of LIGO Scientific Collaboration / Virgo interferometer collaboration presenting at the press conference for the announcement that the first direct gravitational wave observation had been made in September 2015.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, LOCATION, EVENT\n\nText:\nIn February 2016, he was one of the four scientists of LIGO Scientific Collaboration / Virgo interferometer collaboration presenting at the press conference for the announcement that the first direct gravitational wave observation had been made in September 2015.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_7", "text": "Research indicates that the majority of variation in coat growth pattern, length and curl can be attributed to mutations in four genes, the R-spondin-2 gene or RSPO2, the fibroblast growth factor-5 gene or FGF5, the KRT71 gene or KRT71 and the melanocortin 5 receptor gene (MC5R).", "requested_types": ["UNIVERSITY", "DISCIPLINE", "MISC", "PROTEIN"], "distractors": ["DISCIPLINE", "UNIVERSITY"], "gold": {"UNIVERSITY": [], "DISCIPLINE": [], "MISC": ["KRT71 gene", "KRT71"], "PROTEIN": ["R-spondin-2", "RSPO2", "fibroblast growth factor-5", "FGF5", "melanocortin 5 receptor", "MC5R"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, DISCIPLINE, MISC, PROTEIN\n\nText:\nResearch indicates that the majority of variation in coat growth pattern, length and curl can be attributed to mutations in four genes, the R-spondin-2 gene or RSPO2, the fibroblast growth factor-5 gene or FGF5, the KRT71 gene or KRT71 and the melanocortin 5 receptor gene (MC5R).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, DISCIPLINE, MISC, PROTEIN\n\nText:\nResearch indicates that the majority of variation in coat growth pattern, length and curl can be attributed to mutations in four genes, the R-spondin-2 gene or RSPO2, the fibroblast growth factor-5 gene or FGF5, the KRT71 gene or KRT71 and the melanocortin 5 receptor gene (MC5R).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_8", "text": "On 9 May 1821 he married in Klein Lankum with Camper's third child and second daughter Frederika Theodora Ernestina Camper (1799-1834), who was herself an amateur-scientist who accompanied him on voyages to Georges Cuvier in France and Humphry Davy in England; she made drawings of his specimens.", "requested_types": ["LOCATION", "PERSON", "PROTEIN", "EVENT", "COUNTRY", "SCIENTIST"], "distractors": ["PROTEIN", "EVENT"], "gold": {"LOCATION": ["Klein Lankum"], "PERSON": ["Camper", "Frederika Theodora Ernestina Camper"], "PROTEIN": [], "EVENT": [], "COUNTRY": ["France", "England"], "SCIENTIST": ["Georges Cuvier", "Humphry Davy"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, PERSON, PROTEIN, EVENT, COUNTRY, SCIENTIST\n\nText:\nOn 9 May 1821 he married in Klein Lankum with Camper's third child and second daughter Frederika Theodora Ernestina Camper (1799-1834), who was herself an amateur-scientist who accompanied him on voyages to Georges Cuvier in France and Humphry Davy in England; she made drawings of his specimens.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, PERSON, PROTEIN, EVENT, COUNTRY, SCIENTIST\n\nText:\nOn 9 May 1821 he married in Klein Lankum with Camper's third child and second daughter Frederika Theodora Ernestina Camper (1799-1834), who was herself an amateur-scientist who accompanied him on voyages to Georges Cuvier in France and Humphry Davy in England; she made drawings of his specimens.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_9", "text": "That is, cell stimulation causes DHA to be released from the sn-2 position of their membrane-bound cellular phospholipid pools through the action of a Phospholipase A2 -type enzyme and the subsequent attack of the released DHA by CYP450 epoxidases.", "requested_types": ["SCIENTIST", "CHEMICAL_COMPOUND", "MISC", "ENZYME"], "distractors": ["SCIENTIST"], "gold": {"SCIENTIST": [], "CHEMICAL_COMPOUND": ["DHA"], "MISC": ["sn-2 position", "membrane-bound cellular phospholipid pools"], "ENZYME": ["Phospholipase A2 -type enzyme", "CYP450 epoxidases"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_COMPOUND, MISC, ENZYME\n\nText:\nThat is, cell stimulation causes DHA to be released from the sn-2 position of their membrane-bound cellular phospholipid pools through the action of a Phospholipase A2 -type enzyme and the subsequent attack of the released DHA by CYP450 epoxidases.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_COMPOUND, MISC, ENZYME\n\nText:\nThat is, cell stimulation causes DHA to be released from the sn-2 position of their membrane-bound cellular phospholipid pools through the action of a Phospholipase A2 -type enzyme and the subsequent attack of the released DHA by CYP450 epoxidases.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_10", "text": "Football is also a common sport and the multi-use Saifur Rahman Stadium and Netaji Subhas Chandra Bose Stadium s are known to host football matches.", "requested_types": ["LOCATION", "EVENT", "UNIVERSITY"], "distractors": ["UNIVERSITY", "EVENT"], "gold": {"LOCATION": ["Saifur Rahman Stadium", "Netaji Subhas Chandra Bose Stadium"], "EVENT": [], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, EVENT, UNIVERSITY\n\nText:\nFootball is also a common sport and the multi-use Saifur Rahman Stadium and Netaji Subhas Chandra Bose Stadium s are known to host football matches.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, EVENT, UNIVERSITY\n\nText:\nFootball is also a common sport and the multi-use Saifur Rahman Stadium and Netaji Subhas Chandra Bose Stadium s are known to host football matches.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_11", "text": "Khan's research focus is on developing new inhibitors against multidrug resistant clinical strains with special interest on extended spectrum beta lactamases (ESBL) such as NDM-1 and Beta-lactamase, using QSARR modeling and structure-based virtual screening methods.", "requested_types": ["MISC", "LOCATION", "AWARD", "ENZYME", "COUNTRY", "SCIENTIST"], "distractors": ["AWARD", "COUNTRY", "LOCATION"], "gold": {"MISC": ["multidrug resistant", "QSARR"], "LOCATION": [], "AWARD": [], "ENZYME": ["extended spectrum beta lactamases", "ESBL", "NDM-1", "Beta-lactamase"], "COUNTRY": [], "SCIENTIST": ["Khan"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, LOCATION, AWARD, ENZYME, COUNTRY, SCIENTIST\n\nText:\nKhan's research focus is on developing new inhibitors against multidrug resistant clinical strains with special interest on extended spectrum beta lactamases (ESBL) such as NDM-1 and Beta-lactamase, using QSARR modeling and structure-based virtual screening methods.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, LOCATION, AWARD, ENZYME, COUNTRY, SCIENTIST\n\nText:\nKhan's research focus is on developing new inhibitors against multidrug resistant clinical strains with special interest on extended spectrum beta lactamases (ESBL) such as NDM-1 and Beta-lactamase, using QSARR modeling and structure-based virtual screening methods.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_12", "text": "Polyethylene terephthalate (PET) bottles are made from ethylene and P-Xylene.", "requested_types": ["ASTRONOMICAL_OBJECT", "UNIVERSITY", "ACADEMIC_JOURNAL", "CHEMICAL_COMPOUND"], "distractors": ["ASTRONOMICAL_OBJECT", "UNIVERSITY", "ACADEMIC_JOURNAL"], "gold": {"ASTRONOMICAL_OBJECT": [], "UNIVERSITY": [], "ACADEMIC_JOURNAL": [], "CHEMICAL_COMPOUND": ["Polyethylene terephthalate", "PET", "ethylene", "P-Xylene"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, UNIVERSITY, ACADEMIC_JOURNAL, CHEMICAL_COMPOUND\n\nText:\nPolyethylene terephthalate (PET) bottles are made from ethylene and P-Xylene.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, UNIVERSITY, ACADEMIC_JOURNAL, CHEMICAL_COMPOUND\n\nText:\nPolyethylene terephthalate (PET) bottles are made from ethylene and P-Xylene.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_13", "text": "Most widely used to assess potential synthetic lethal interactions is using siRNA and CRISPR to modify target genes.", "requested_types": ["ORGANISATION", "EVENT", "MISC"], "distractors": ["ORGANISATION", "EVENT"], "gold": {"ORGANISATION": [], "EVENT": [], "MISC": ["siRNA", "CRISPR"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, EVENT, MISC\n\nText:\nMost widely used to assess potential synthetic lethal interactions is using siRNA and CRISPR to modify target genes.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, EVENT, MISC\n\nText:\nMost widely used to assess potential synthetic lethal interactions is using siRNA and CRISPR to modify target genes.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_14", "text": "In the low pressure process α-olefins (e.g. 1-Butene or 1-Hexene) may be added, which are incorporated in the polymer chain during polymerization.", "requested_types": ["LOCATION", "CHEMICAL_COMPOUND"], "distractors": ["LOCATION"], "gold": {"LOCATION": [], "CHEMICAL_COMPOUND": ["α-olefins", "1-Butene", "1-Hexene"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, CHEMICAL_COMPOUND\n\nText:\nIn the low pressure process α-olefins (e.g. 1-Butene or 1-Hexene) may be added, which are incorporated in the polymer chain during polymerization.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, CHEMICAL_COMPOUND\n\nText:\nIn the low pressure process α-olefins (e.g. 1-Butene or 1-Hexene) may be added, which are incorporated in the polymer chain during polymerization.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_15", "text": "Cory's work has been published in research journals including Blood (journal), The EMBO Journal, Nature (journal), Cell Death & amp; Differentiation, and Proceedings of the National Academy of Sciences of the United States of America.", "requested_types": ["ACADEMIC_JOURNAL", "EVENT", "SCIENTIST", "ASTRONOMICAL_OBJECT"], "distractors": ["ASTRONOMICAL_OBJECT", "EVENT"], "gold": {"ACADEMIC_JOURNAL": ["Blood", "The EMBO Journal", "Nature", "Cell Death & amp; Differentiation", "Proceedings of the National Academy of Sciences of the United States of America"], "EVENT": [], "SCIENTIST": ["Cory"], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, EVENT, SCIENTIST, ASTRONOMICAL_OBJECT\n\nText:\nCory's work has been published in research journals including Blood (journal), The EMBO Journal, Nature (journal), Cell Death & amp; Differentiation, and Proceedings of the National Academy of Sciences of the United States of America.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, EVENT, SCIENTIST, ASTRONOMICAL_OBJECT\n\nText:\nCory's work has been published in research journals including Blood (journal), The EMBO Journal, Nature (journal), Cell Death & amp; Differentiation, and Proceedings of the National Academy of Sciences of the United States of America.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_16", "text": "Albarn teamed up with Robert 3D Del Naja of Massive Attack and worked with Stop the War Coalition, CND and the Muslim Association of Britain to organise campaigns to raise awareness of the potential dangers of the UK's involvement in the war.", "requested_types": ["ORGANISATION", "COUNTRY", "AWARD", "THEORY", "DISCIPLINE", "PERSON"], "distractors": ["DISCIPLINE", "AWARD", "THEORY"], "gold": {"ORGANISATION": ["Massive Attack", "Stop the War Coalition", "CND", "Muslim Association of Britain"], "COUNTRY": ["UK"], "AWARD": [], "THEORY": [], "DISCIPLINE": [], "PERSON": ["Albarn", "Robert 3D Del Naja"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, COUNTRY, AWARD, THEORY, DISCIPLINE, PERSON\n\nText:\nAlbarn teamed up with Robert 3D Del Naja of Massive Attack and worked with Stop the War Coalition, CND and the Muslim Association of Britain to organise campaigns to raise awareness of the potential dangers of the UK's involvement in the war.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, COUNTRY, AWARD, THEORY, DISCIPLINE, PERSON\n\nText:\nAlbarn teamed up with Robert 3D Del Naja of Massive Attack and worked with Stop the War Coalition, CND and the Muslim Association of Britain to organise campaigns to raise awareness of the potential dangers of the UK's involvement in the war.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_17", "text": "Further articles appeared on neuroethics in Nature Neuroscience, Neuron, and Brain and Cognition.", "requested_types": ["ORGANISATION", "ACADEMIC_JOURNAL"], "distractors": ["ORGANISATION"], "gold": {"ORGANISATION": [], "ACADEMIC_JOURNAL": ["Nature Neuroscience", "Neuron", "Brain and Cognition"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ACADEMIC_JOURNAL\n\nText:\nFurther articles appeared on neuroethics in Nature Neuroscience, Neuron, and Brain and Cognition.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ACADEMIC_JOURNAL\n\nText:\nFurther articles appeared on neuroethics in Nature Neuroscience, Neuron, and Brain and Cognition.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_18", "text": "The primary effectors of Gβγ are various ion channels, such as G protein-coupled inwardly-rectifying potassium channel (GIRKs), P-type calcium channel / Q-type calcium channel - and N- type voltage-gated Casup2 + / sup channels, as well as some isoforms of AC and PLC, along with some phosphoinositide-3-kinase (PI3K) isoforms.", "requested_types": ["COUNTRY", "MISC", "ORGANISATION", "PROTEIN", "ENZYME"], "distractors": ["COUNTRY", "ORGANISATION"], "gold": {"COUNTRY": [], "MISC": ["ion channels", "G protein-coupled inwardly-rectifying potassium channel", "GIRKs", "P-type calcium channel", "Q-type calcium channel", "N- type voltage-gated", "Casup2 + / sup"], "ORGANISATION": [], "PROTEIN": ["Gβγ"], "ENZYME": ["AC", "PLC", "phosphoinositide-3-kinase", "PI3K"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, MISC, ORGANISATION, PROTEIN, ENZYME\n\nText:\nThe primary effectors of Gβγ are various ion channels, such as G protein-coupled inwardly-rectifying potassium channel (GIRKs), P-type calcium channel / Q-type calcium channel - and N- type voltage-gated Casup2 + / sup channels, as well as some isoforms of AC and PLC, along with some phosphoinositide-3-kinase (PI3K) isoforms.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, MISC, ORGANISATION, PROTEIN, ENZYME\n\nText:\nThe primary effectors of Gβγ are various ion channels, such as G protein-coupled inwardly-rectifying potassium channel (GIRKs), P-type calcium channel / Q-type calcium channel - and N- type voltage-gated Casup2 + / sup channels, as well as some isoforms of AC and PLC, along with some phosphoinositide-3-kinase (PI3K) isoforms.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_19", "text": "It displays the mean time, sidereal, (or star), time and the motions of the sun, moon and the five then-known planets Venus, Mars, Saturn, Mercury, and Jupiter.", "requested_types": ["ASTRONOMICAL_OBJECT", "MISC", "SCIENTIST"], "distractors": ["MISC", "SCIENTIST"], "gold": {"ASTRONOMICAL_OBJECT": ["sun", "moon", "Venus", "Mars", "Saturn", "Mercury", "Jupiter"], "MISC": [], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, SCIENTIST\n\nText:\nIt displays the mean time, sidereal, (or star), time and the motions of the sun, moon and the five then-known planets Venus, Mars, Saturn, Mercury, and Jupiter.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, SCIENTIST\n\nText:\nIt displays the mean time, sidereal, (or star), time and the motions of the sun, moon and the five then-known planets Venus, Mars, Saturn, Mercury, and Jupiter.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_20", "text": "Hamilton was a visiting professor at Harvard University and later spent nine months with the Royal Society 's and the Royal Geographical Society' s Xavantina-Cachimbo Expedition as a visiting professor at the University of São Paulo.", "requested_types": ["EVENT", "ORGANISATION", "MISC", "UNIVERSITY", "SCIENTIST"], "distractors": ["MISC"], "gold": {"EVENT": ["Xavantina-Cachimbo Expedition"], "ORGANISATION": ["Royal Society", "Royal Geographical Society"], "MISC": [], "UNIVERSITY": ["Harvard University", "University of São Paulo"], "SCIENTIST": ["Hamilton"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, ORGANISATION, MISC, UNIVERSITY, SCIENTIST\n\nText:\nHamilton was a visiting professor at Harvard University and later spent nine months with the Royal Society 's and the Royal Geographical Society' s Xavantina-Cachimbo Expedition as a visiting professor at the University of São Paulo.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, ORGANISATION, MISC, UNIVERSITY, SCIENTIST\n\nText:\nHamilton was a visiting professor at Harvard University and later spent nine months with the Royal Society 's and the Royal Geographical Society' s Xavantina-Cachimbo Expedition as a visiting professor at the University of São Paulo.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_21", "text": "In the second season, married couple Rose Nadler (L. Scott Caldwell) and Bernard Nadler (Sam Anderson), separated on opposite sides of the island (she with the main characters, he with the tail section survivors), were featured in a flashback episode after being reunited.", "requested_types": ["DISCIPLINE", "AWARD", "PERSON"], "distractors": ["DISCIPLINE", "AWARD"], "gold": {"DISCIPLINE": [], "AWARD": [], "PERSON": ["Rose Nadler", "L. Scott Caldwell", "Bernard Nadler", "Sam Anderson"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, AWARD, PERSON\n\nText:\nIn the second season, married couple Rose Nadler (L. Scott Caldwell) and Bernard Nadler (Sam Anderson), separated on opposite sides of the island (she with the main characters, he with the tail section survivors), were featured in a flashback episode after being reunited.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, AWARD, PERSON\n\nText:\nIn the second season, married couple Rose Nadler (L. Scott Caldwell) and Bernard Nadler (Sam Anderson), separated on opposite sides of the island (she with the main characters, he with the tail section survivors), were featured in a flashback episode after being reunited.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_22", "text": "In the presence of Casup2 + / sup, calmodulin and S100 protein bind calponin at the region of actin-binding sites and reverse calponin's inhibition of myosin MgATPase.", "requested_types": ["ENZYME", "UNIVERSITY", "PROTEIN", "CHEMICAL_ELEMENT", "LOCATION"], "distractors": ["LOCATION", "UNIVERSITY"], "gold": {"ENZYME": ["myosin MgATPase"], "UNIVERSITY": [], "PROTEIN": ["calmodulin", "S100 protein", "calponin"], "CHEMICAL_ELEMENT": ["Casup2 + / sup"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, UNIVERSITY, PROTEIN, CHEMICAL_ELEMENT, LOCATION\n\nText:\nIn the presence of Casup2 + / sup, calmodulin and S100 protein bind calponin at the region of actin-binding sites and reverse calponin's inhibition of myosin MgATPase.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, UNIVERSITY, PROTEIN, CHEMICAL_ELEMENT, LOCATION\n\nText:\nIn the presence of Casup2 + / sup, calmodulin and S100 protein bind calponin at the region of actin-binding sites and reverse calponin's inhibition of myosin MgATPase.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_23", "text": "The amplified fragment (amplicon) is incubated with the restriction enzyme HaeIII, comprising the SNP in their recognition sequence GGCC.", "requested_types": ["MISC", "LOCATION", "CHEMICAL_ELEMENT", "ENZYME", "ORGANISATION"], "distractors": ["ORGANISATION", "LOCATION", "CHEMICAL_ELEMENT"], "gold": {"MISC": ["amplicon", "SNP"], "LOCATION": [], "CHEMICAL_ELEMENT": [], "ENZYME": ["HaeIII"], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, LOCATION, CHEMICAL_ELEMENT, ENZYME, ORGANISATION\n\nText:\nThe amplified fragment (amplicon) is incubated with the restriction enzyme HaeIII, comprising the SNP in their recognition sequence GGCC.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, LOCATION, CHEMICAL_ELEMENT, ENZYME, ORGANISATION\n\nText:\nThe amplified fragment (amplicon) is incubated with the restriction enzyme HaeIII, comprising the SNP in their recognition sequence GGCC.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_24", "text": "She has won five Grammy Awards, including Grammy Award for Album of the Year and Grammy Award for Record of the Year.", "requested_types": ["AWARD", "THEORY", "PROTEIN", "SCIENTIST"], "distractors": ["THEORY", "SCIENTIST", "PROTEIN"], "gold": {"AWARD": ["Grammy Awards", "Grammy Award for Album of the Year", "Grammy Award for Record of the Year"], "THEORY": [], "PROTEIN": [], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, THEORY, PROTEIN, SCIENTIST\n\nText:\nShe has won five Grammy Awards, including Grammy Award for Album of the Year and Grammy Award for Record of the Year.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, THEORY, PROTEIN, SCIENTIST\n\nText:\nShe has won five Grammy Awards, including Grammy Award for Album of the Year and Grammy Award for Record of the Year.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_25", "text": "Some asteroid s, also named after the same Shakespearean characters, share names with moons of Uranus: 171 Ophelia, 218 Bianca, 593 Titania, 666 Desdemona, 763 Cupido, and 2758 Cordelia.", "requested_types": ["ASTRONOMICAL_OBJECT", "DISCIPLINE"], "distractors": ["DISCIPLINE"], "gold": {"ASTRONOMICAL_OBJECT": ["Uranus", "171 Ophelia", "218 Bianca", "593 Titania", "666 Desdemona", "763 Cupido", "2758 Cordelia"], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, DISCIPLINE\n\nText:\nSome asteroid s, also named after the same Shakespearean characters, share names with moons of Uranus: 171 Ophelia, 218 Bianca, 593 Titania, 666 Desdemona, 763 Cupido, and 2758 Cordelia.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, DISCIPLINE\n\nText:\nSome asteroid s, also named after the same Shakespearean characters, share names with moons of Uranus: 171 Ophelia, 218 Bianca, 593 Titania, 666 Desdemona, 763 Cupido, and 2758 Cordelia.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_26", "text": "It was opened in 1982 to host the 1982 Southern Cross Games and since then it has hosted the 1983 South American Championships in Athletics; the 1985, 1994 and 2001 South American Junior Championships in Athletics; and the 1989 and 2001 Pan American Junior Athletics Championships, beside numerous seminars, congresses and courses.", "requested_types": ["CHEMICAL_COMPOUND", "EVENT", "MISC", "ACADEMIC_JOURNAL", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_ELEMENT", "CHEMICAL_COMPOUND", "ACADEMIC_JOURNAL"], "gold": {"CHEMICAL_COMPOUND": [], "EVENT": ["1982 Southern Cross Games", "1983 South American Championships", "2001 South American Junior Championships", "2001 Pan American Junior Athletics Championships"], "MISC": ["Athletics"], "ACADEMIC_JOURNAL": [], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, EVENT, MISC, ACADEMIC_JOURNAL, CHEMICAL_ELEMENT\n\nText:\nIt was opened in 1982 to host the 1982 Southern Cross Games and since then it has hosted the 1983 South American Championships in Athletics; the 1985, 1994 and 2001 South American Junior Championships in Athletics; and the 1989 and 2001 Pan American Junior Athletics Championships, beside numerous seminars, congresses and courses.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, EVENT, MISC, ACADEMIC_JOURNAL, CHEMICAL_ELEMENT\n\nText:\nIt was opened in 1982 to host the 1982 Southern Cross Games and since then it has hosted the 1983 South American Championships in Athletics; the 1985, 1994 and 2001 South American Junior Championships in Athletics; and the 1989 and 2001 Pan American Junior Athletics Championships, beside numerous seminars, congresses and courses.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_27", "text": "GAR + 10-Formyltetrahydrofolate → Phosphoribosyl-N-formylglycineamide + Tetrahydrofolic acid", "requested_types": ["SCIENTIST", "UNIVERSITY", "CHEMICAL_COMPOUND"], "distractors": ["UNIVERSITY", "SCIENTIST"], "gold": {"SCIENTIST": [], "UNIVERSITY": [], "CHEMICAL_COMPOUND": ["GAR", "10-Formyltetrahydrofolate", "Phosphoribosyl-N-formylglycineamide", "Tetrahydrofolic acid"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, UNIVERSITY, CHEMICAL_COMPOUND\n\nText:\nGAR + 10-Formyltetrahydrofolate → Phosphoribosyl-N-formylglycineamide + Tetrahydrofolic acid\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, UNIVERSITY, CHEMICAL_COMPOUND\n\nText:\nGAR + 10-Formyltetrahydrofolate → Phosphoribosyl-N-formylglycineamide + Tetrahydrofolic acid\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_28", "text": "Alferov and colleagues worked on Gallium arsenide and Aluminium arsenide III-V heterojunctions.", "requested_types": ["EVENT", "MISC", "CHEMICAL_COMPOUND", "COUNTRY", "SCIENTIST"], "distractors": ["EVENT", "COUNTRY", "MISC"], "gold": {"EVENT": [], "MISC": [], "CHEMICAL_COMPOUND": ["Gallium arsenide", "Aluminium arsenide"], "COUNTRY": [], "SCIENTIST": ["Alferov"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, MISC, CHEMICAL_COMPOUND, COUNTRY, SCIENTIST\n\nText:\nAlferov and colleagues worked on Gallium arsenide and Aluminium arsenide III-V heterojunctions.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, MISC, CHEMICAL_COMPOUND, COUNTRY, SCIENTIST\n\nText:\nAlferov and colleagues worked on Gallium arsenide and Aluminium arsenide III-V heterojunctions.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_29", "text": "She has listed Christina Aguilera, Jessica Simpson, Bobbie Gentry, Janis Joplin, Shelby Lynne, and Shania Twain as her musical influences.", "requested_types": ["ACADEMIC_JOURNAL", "PERSON", "CHEMICAL_COMPOUND", "MISC"], "distractors": ["ACADEMIC_JOURNAL", "MISC", "CHEMICAL_COMPOUND"], "gold": {"ACADEMIC_JOURNAL": [], "PERSON": ["Christina Aguilera", "Jessica Simpson", "Bobbie Gentry", "Janis Joplin", "Shelby Lynne", "Shania Twain"], "CHEMICAL_COMPOUND": [], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, PERSON, CHEMICAL_COMPOUND, MISC\n\nText:\nShe has listed Christina Aguilera, Jessica Simpson, Bobbie Gentry, Janis Joplin, Shelby Lynne, and Shania Twain as her musical influences.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, PERSON, CHEMICAL_COMPOUND, MISC\n\nText:\nShe has listed Christina Aguilera, Jessica Simpson, Bobbie Gentry, Janis Joplin, Shelby Lynne, and Shania Twain as her musical influences.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_30", "text": "The Zeitschrift für Naturforschung (English: Journal for Nature Research) was established in 1946 by the Max Planck Institute and the physical sciences (Zeitschrift für Naturforschung A) were separated from the other natural sciences (Part B) from 1947.", "requested_types": ["PERSON", "ACADEMIC_JOURNAL", "ASTRONOMICAL_OBJECT", "ORGANISATION", "MISC", "EVENT"], "distractors": ["ASTRONOMICAL_OBJECT", "PERSON", "EVENT"], "gold": {"PERSON": [], "ACADEMIC_JOURNAL": ["Zeitschrift für Naturforschung", "Journal for Nature Research", "the physical sciences (Zeitschrift für Naturforschung A)"], "ASTRONOMICAL_OBJECT": [], "ORGANISATION": ["Max Planck Institute"], "MISC": ["English"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ACADEMIC_JOURNAL, ASTRONOMICAL_OBJECT, ORGANISATION, MISC, EVENT\n\nText:\nThe Zeitschrift für Naturforschung (English: Journal for Nature Research) was established in 1946 by the Max Planck Institute and the physical sciences (Zeitschrift für Naturforschung A) were separated from the other natural sciences (Part B) from 1947.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ACADEMIC_JOURNAL, ASTRONOMICAL_OBJECT, ORGANISATION, MISC, EVENT\n\nText:\nThe Zeitschrift für Naturforschung (English: Journal for Nature Research) was established in 1946 by the Max Planck Institute and the physical sciences (Zeitschrift für Naturforschung A) were separated from the other natural sciences (Part B) from 1947.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_31", "text": "A second meeting was held soon thereafter and included Klaus Clusius, Robert Döpel, Werner Heisenberg, and Carl Friedrich von Weizsäcker.", "requested_types": ["PROTEIN", "CHEMICAL_COMPOUND", "MISC", "SCIENTIST"], "distractors": ["CHEMICAL_COMPOUND", "MISC", "PROTEIN"], "gold": {"PROTEIN": [], "CHEMICAL_COMPOUND": [], "MISC": [], "SCIENTIST": ["Klaus Clusius", "Robert Döpel", "Werner Heisenberg", "Carl Friedrich von Weizsäcker"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_COMPOUND, MISC, SCIENTIST\n\nText:\nA second meeting was held soon thereafter and included Klaus Clusius, Robert Döpel, Werner Heisenberg, and Carl Friedrich von Weizsäcker.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_COMPOUND, MISC, SCIENTIST\n\nText:\nA second meeting was held soon thereafter and included Klaus Clusius, Robert Döpel, Werner Heisenberg, and Carl Friedrich von Weizsäcker.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_32", "text": "He is the recipient of a Lifetime Achievement Award from the The Recording Academy and a Lifetime Achievement Award from the Rhythm and Blues Foundation.", "requested_types": ["AWARD", "THEORY", "ASTRONOMICAL_OBJECT", "MISC"], "distractors": ["THEORY", "MISC", "ASTRONOMICAL_OBJECT"], "gold": {"AWARD": ["Lifetime Achievement Award from the The Recording Academy", "Lifetime Achievement Award from the Rhythm and Blues Foundation"], "THEORY": [], "ASTRONOMICAL_OBJECT": [], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, THEORY, ASTRONOMICAL_OBJECT, MISC\n\nText:\nHe is the recipient of a Lifetime Achievement Award from the The Recording Academy and a Lifetime Achievement Award from the Rhythm and Blues Foundation.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, THEORY, ASTRONOMICAL_OBJECT, MISC\n\nText:\nHe is the recipient of a Lifetime Achievement Award from the The Recording Academy and a Lifetime Achievement Award from the Rhythm and Blues Foundation.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_33", "text": "Self-harmful T cells, further referred to as autoreactive T cells, originate in the thymus because of the stochastic process called V (D) J recombination which conducts the generation of T-cell receptor (TCRs) and enables their limitless variability.", "requested_types": ["PROTEIN", "AWARD", "MISC", "ORGANISATION"], "distractors": ["ORGANISATION", "AWARD"], "gold": {"PROTEIN": ["T-cell receptor", "TCRs"], "AWARD": [], "MISC": ["Self-harmful T cells", "autoreactive T cells", "thymus", "V (D) J recombination"], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, AWARD, MISC, ORGANISATION\n\nText:\nSelf-harmful T cells, further referred to as autoreactive T cells, originate in the thymus because of the stochastic process called V (D) J recombination which conducts the generation of T-cell receptor (TCRs) and enables their limitless variability.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, AWARD, MISC, ORGANISATION\n\nText:\nSelf-harmful T cells, further referred to as autoreactive T cells, originate in the thymus because of the stochastic process called V (D) J recombination which conducts the generation of T-cell receptor (TCRs) and enables their limitless variability.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_34", "text": "He has chaired the Southern Growth Policies Board, the Southern Regional Education Board, the Southern Technology Council, the Interstate Oil and Gas Compact Commission, and the Education Commission of the States.", "requested_types": ["UNIVERSITY", "ORGANISATION", "MISC"], "distractors": ["MISC", "UNIVERSITY"], "gold": {"UNIVERSITY": [], "ORGANISATION": ["Southern Growth Policies Board", "Southern Regional Education Board", "the Southern Technology Council", "Interstate Oil and Gas Compact Commission", "Education Commission of the States"], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ORGANISATION, MISC\n\nText:\nHe has chaired the Southern Growth Policies Board, the Southern Regional Education Board, the Southern Technology Council, the Interstate Oil and Gas Compact Commission, and the Education Commission of the States.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ORGANISATION, MISC\n\nText:\nHe has chaired the Southern Growth Policies Board, the Southern Regional Education Board, the Southern Technology Council, the Interstate Oil and Gas Compact Commission, and the Education Commission of the States.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_35", "text": "The specific ligand of this receptor is CCL25 To note, the chemokine binding protein D6 had previously been named CCR9, but this molecule is a scavenger receptor not a TRUE (signaling) chemokine receptor.", "requested_types": ["MISC", "EVENT", "PERSON", "PROTEIN", "AWARD"], "distractors": ["EVENT", "PERSON", "AWARD"], "gold": {"MISC": ["scavenger receptor", "chemokine receptor"], "EVENT": [], "PERSON": [], "PROTEIN": ["CCL25", "D6", "CCR9"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, EVENT, PERSON, PROTEIN, AWARD\n\nText:\nThe specific ligand of this receptor is CCL25 To note, the chemokine binding protein D6 had previously been named CCR9, but this molecule is a scavenger receptor not a TRUE (signaling) chemokine receptor.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, EVENT, PERSON, PROTEIN, AWARD\n\nText:\nThe specific ligand of this receptor is CCL25 To note, the chemokine binding protein D6 had previously been named CCR9, but this molecule is a scavenger receptor not a TRUE (signaling) chemokine receptor.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_36", "text": "This type of oxidative stress causes a loss of NF-κB signaling.", "requested_types": ["MISC", "PROTEIN", "ENZYME", "UNIVERSITY"], "distractors": ["UNIVERSITY", "ENZYME"], "gold": {"MISC": ["oxidative stress"], "PROTEIN": ["NF-κB"], "ENZYME": [], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, ENZYME, UNIVERSITY\n\nText:\nThis type of oxidative stress causes a loss of NF-κB signaling.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, ENZYME, UNIVERSITY\n\nText:\nThis type of oxidative stress causes a loss of NF-κB signaling.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_37", "text": "Mikhail Rabinovich published over 250 peer reviewed articles in leading scientific journals (Science, Neuron, Journal of Neuroscience, PLoS Computational Biology, Reviews of Modern Physics, Physics of Fluids, Physical Review Letters) that are actively cited (over 3000 citations).", "requested_types": ["COUNTRY", "ACADEMIC_JOURNAL", "SCIENTIST"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "ACADEMIC_JOURNAL": ["Science", "Neuron", "Journal of Neuroscience", "PLoS Computational Biology", "Reviews of Modern Physics", "Physics of Fluids", "Physical Review Letters"], "SCIENTIST": ["Mikhail Rabinovich"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ACADEMIC_JOURNAL, SCIENTIST\n\nText:\nMikhail Rabinovich published over 250 peer reviewed articles in leading scientific journals (Science, Neuron, Journal of Neuroscience, PLoS Computational Biology, Reviews of Modern Physics, Physics of Fluids, Physical Review Letters) that are actively cited (over 3000 citations).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ACADEMIC_JOURNAL, SCIENTIST\n\nText:\nMikhail Rabinovich published over 250 peer reviewed articles in leading scientific journals (Science, Neuron, Journal of Neuroscience, PLoS Computational Biology, Reviews of Modern Physics, Physics of Fluids, Physical Review Letters) that are actively cited (over 3000 citations).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_38", "text": "Currently, three major groups of DNA binding proteins have been predominantly used for epigenome editing: Zinc finger proteins, Transcription Activator-Like Effectors (TALEs) and nuclease deficient Cas9 fusions (CRISPR).", "requested_types": ["PROTEIN", "COUNTRY", "MISC"], "distractors": ["COUNTRY"], "gold": {"PROTEIN": ["Zinc finger proteins", "Transcription Activator-Like Effectors", "TALEs", "Cas9", "CRISPR"], "COUNTRY": [], "MISC": ["DNA"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, COUNTRY, MISC\n\nText:\nCurrently, three major groups of DNA binding proteins have been predominantly used for epigenome editing: Zinc finger proteins, Transcription Activator-Like Effectors (TALEs) and nuclease deficient Cas9 fusions (CRISPR).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, COUNTRY, MISC\n\nText:\nCurrently, three major groups of DNA binding proteins have been predominantly used for epigenome editing: Zinc finger proteins, Transcription Activator-Like Effectors (TALEs) and nuclease deficient Cas9 fusions (CRISPR).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_39", "text": "On screen she has starred in scores of films and is a six-time Oscar nominee, winning the Academy Award for Best Supporting Actress for the title role in the film Julia (1977). 1977).", "requested_types": ["PROTEIN", "AWARD", "MISC"], "distractors": ["PROTEIN"], "gold": {"PROTEIN": [], "AWARD": ["Oscar", "Academy Award for Best Supporting Actress"], "MISC": ["Julia"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, AWARD, MISC\n\nText:\nOn screen she has starred in scores of films and is a six-time Oscar nominee, winning the Academy Award for Best Supporting Actress for the title role in the film Julia (1977). 1977).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, AWARD, MISC\n\nText:\nOn screen she has starred in scores of films and is a six-time Oscar nominee, winning the Academy Award for Best Supporting Actress for the title role in the film Julia (1977). 1977).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_40", "text": "By 1914, experiments by physicists Ernest Rutherford, Henry Moseley, James Franck and Gustav Ludwig Hertz had largely established the structure of an atom as a dense nucleus of positive charge surrounded by lower-mass electrons.", "requested_types": ["MISC", "SCIENTIST"], "distractors": ["MISC"], "gold": {"MISC": [], "SCIENTIST": ["Ernest Rutherford", "Henry Moseley", "James Franck", "Gustav Ludwig Hertz"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST\n\nText:\nBy 1914, experiments by physicists Ernest Rutherford, Henry Moseley, James Franck and Gustav Ludwig Hertz had largely established the structure of an atom as a dense nucleus of positive charge surrounded by lower-mass electrons.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST\n\nText:\nBy 1914, experiments by physicists Ernest Rutherford, Henry Moseley, James Franck and Gustav Ludwig Hertz had largely established the structure of an atom as a dense nucleus of positive charge surrounded by lower-mass electrons.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_41", "text": "When the discovery of a new element is claimed, the Joint Working Party (JWP) of the International Union of Pure and Applied Chemistry (IUPAC) and the International Union of Pure and Applied Physics (IUPAP) assembles to examine the claims according to their criteria for the discovery of a new element, and decides scientific priority and naming rights for the elements.", "requested_types": ["ASTRONOMICAL_OBJECT", "LOCATION", "ORGANISATION"], "distractors": ["LOCATION", "ASTRONOMICAL_OBJECT"], "gold": {"ASTRONOMICAL_OBJECT": [], "LOCATION": [], "ORGANISATION": ["Joint Working Party", "JWP", "International Union of Pure and Applied Chemistry", "IUPAC", "International Union of Pure and Applied Physics", "IUPAP"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, LOCATION, ORGANISATION\n\nText:\nWhen the discovery of a new element is claimed, the Joint Working Party (JWP) of the International Union of Pure and Applied Chemistry (IUPAC) and the International Union of Pure and Applied Physics (IUPAP) assembles to examine the claims according to their criteria for the discovery of a new element, and decides scientific priority and naming rights for the elements.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, LOCATION, ORGANISATION\n\nText:\nWhen the discovery of a new element is claimed, the Joint Working Party (JWP) of the International Union of Pure and Applied Chemistry (IUPAC) and the International Union of Pure and Applied Physics (IUPAP) assembles to examine the claims according to their criteria for the discovery of a new element, and decides scientific priority and naming rights for the elements.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_42", "text": "He was a Fellow of the Royal Society and a foreign member of the Royal Swedish Academy of Sciences.", "requested_types": ["ORGANISATION", "SCIENTIST", "AWARD"], "distractors": ["SCIENTIST"], "gold": {"ORGANISATION": ["Royal Swedish Academy of Sciences"], "SCIENTIST": [], "AWARD": ["Fellow of the Royal Society"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, SCIENTIST, AWARD\n\nText:\nHe was a Fellow of the Royal Society and a foreign member of the Royal Swedish Academy of Sciences.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, SCIENTIST, AWARD\n\nText:\nHe was a Fellow of the Royal Society and a foreign member of the Royal Swedish Academy of Sciences.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_43", "text": "Further story arcs which provided a particular focus on individual characters include Addison Montgomery-Shepherd (Kate Walsh) facing Derek Shepherd 's (Patrick Dempsey) desire to divorce her, Izzie Stevens (Katherine Heigl) coping with Denny Duquette' s (Jeffrey Dean Morgan) death and the repercussions of her choice to quit the internship program, and Meredith Grey (Ellen Pompeo) struggling with the choice between Shepherd and Finn Dandridge (Chris O'Donnell).", "requested_types": ["PROTEIN", "DISCIPLINE", "PERSON"], "distractors": ["PROTEIN", "DISCIPLINE"], "gold": {"PROTEIN": [], "DISCIPLINE": [], "PERSON": ["Addison Montgomery-Shepherd", "Kate Walsh", "Derek Shepherd", "Patrick Dempsey", "Izzie Stevens", "Katherine Heigl", "Denny Duquette", "Jeffrey Dean Morgan", "Meredith Grey", "Ellen Pompeo", "Shepherd", "Finn Dandridge", "Chris O'Donnell"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, DISCIPLINE, PERSON\n\nText:\nFurther story arcs which provided a particular focus on individual characters include Addison Montgomery-Shepherd (Kate Walsh) facing Derek Shepherd 's (Patrick Dempsey) desire to divorce her, Izzie Stevens (Katherine Heigl) coping with Denny Duquette' s (Jeffrey Dean Morgan) death and the repercussions of her choice to quit the internship program, and Meredith Grey (Ellen Pompeo) struggling with the choice between Shepherd and Finn Dandridge (Chris O'Donnell).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, DISCIPLINE, PERSON\n\nText:\nFurther story arcs which provided a particular focus on individual characters include Addison Montgomery-Shepherd (Kate Walsh) facing Derek Shepherd 's (Patrick Dempsey) desire to divorce her, Izzie Stevens (Katherine Heigl) coping with Denny Duquette' s (Jeffrey Dean Morgan) death and the repercussions of her choice to quit the internship program, and Meredith Grey (Ellen Pompeo) struggling with the choice between Shepherd and Finn Dandridge (Chris O'Donnell).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_44", "text": "Non-governmental agencies such as the International Planned Parenthood Federation and Marie Stopes International provide contraceptive advice for young women worldwide.", "requested_types": ["CHEMICAL_ELEMENT", "ORGANISATION"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_ELEMENT": [], "ORGANISATION": ["International Planned Parenthood Federation", "Marie Stopes International"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, ORGANISATION\n\nText:\nNon-governmental agencies such as the International Planned Parenthood Federation and Marie Stopes International provide contraceptive advice for young women worldwide.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, ORGANISATION\n\nText:\nNon-governmental agencies such as the International Planned Parenthood Federation and Marie Stopes International provide contraceptive advice for young women worldwide.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_45", "text": "Together, they activate multiple signal transduction pathways including oxidant stress, tyrosine kinase s, Protein kinase C, and MAPK s leading to activation of transcription factors such as NF-κB, and dysregulation of epigenetic mechanisms including HKme, histone lysine acetylation, and DNA methylation via the action of corresponding methyltransferases, demethylases, acetylases, and deacetylases.", "requested_types": ["ENZYME", "ORGANISATION", "PROTEIN", "COUNTRY", "MISC", "CHEMICAL_COMPOUND"], "distractors": ["ORGANISATION", "CHEMICAL_COMPOUND", "COUNTRY"], "gold": {"ENZYME": ["tyrosine kinase", "Protein kinase C", "MAPK", "methyltransferases", "demethylases", "acetylases", "deacetylases"], "ORGANISATION": [], "PROTEIN": ["NF-κB"], "COUNTRY": [], "MISC": ["HKme", "histone lysine acetylation", "DNA methylation"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ORGANISATION, PROTEIN, COUNTRY, MISC, CHEMICAL_COMPOUND\n\nText:\nTogether, they activate multiple signal transduction pathways including oxidant stress, tyrosine kinase s, Protein kinase C, and MAPK s leading to activation of transcription factors such as NF-κB, and dysregulation of epigenetic mechanisms including HKme, histone lysine acetylation, and DNA methylation via the action of corresponding methyltransferases, demethylases, acetylases, and deacetylases.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ORGANISATION, PROTEIN, COUNTRY, MISC, CHEMICAL_COMPOUND\n\nText:\nTogether, they activate multiple signal transduction pathways including oxidant stress, tyrosine kinase s, Protein kinase C, and MAPK s leading to activation of transcription factors such as NF-κB, and dysregulation of epigenetic mechanisms including HKme, histone lysine acetylation, and DNA methylation via the action of corresponding methyltransferases, demethylases, acetylases, and deacetylases.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_46", "text": "Young's work on squid giant axons was utilized by Andrew Huxley and Alan Hodgkin who in 1963 received the Nobel Prize for their work on the conduction of action potentials along nerve fibres.", "requested_types": ["ENZYME", "EVENT", "AWARD", "CHEMICAL_COMPOUND", "MISC", "SCIENTIST"], "distractors": ["CHEMICAL_COMPOUND", "ENZYME", "EVENT"], "gold": {"ENZYME": [], "EVENT": [], "AWARD": ["Nobel Prize"], "CHEMICAL_COMPOUND": [], "MISC": ["squid giant axons", "nerve fibres"], "SCIENTIST": ["Young", "Andrew Huxley", "Alan Hodgkin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, EVENT, AWARD, CHEMICAL_COMPOUND, MISC, SCIENTIST\n\nText:\nYoung's work on squid giant axons was utilized by Andrew Huxley and Alan Hodgkin who in 1963 received the Nobel Prize for their work on the conduction of action potentials along nerve fibres.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, EVENT, AWARD, CHEMICAL_COMPOUND, MISC, SCIENTIST\n\nText:\nYoung's work on squid giant axons was utilized by Andrew Huxley and Alan Hodgkin who in 1963 received the Nobel Prize for their work on the conduction of action potentials along nerve fibres.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_47", "text": "The group included the physicists Walther Bothe, Robert Döpel, Hans Geiger, Wolfgang Gentner (probably sent by Walther Bothe), Wilhelm Hanle, Gerhard Hoffmann, and Georg Joos; Peter Debye was invited, but he did not attend.", "requested_types": ["EVENT", "CHEMICAL_ELEMENT", "SCIENTIST"], "distractors": ["EVENT", "CHEMICAL_ELEMENT"], "gold": {"EVENT": [], "CHEMICAL_ELEMENT": [], "SCIENTIST": ["Walther Bothe", "Robert Döpel", "Hans Geiger", "Wolfgang Gentner", "Wilhelm Hanle", "Gerhard Hoffmann", "Georg Joos", "Peter Debye"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, CHEMICAL_ELEMENT, SCIENTIST\n\nText:\nThe group included the physicists Walther Bothe, Robert Döpel, Hans Geiger, Wolfgang Gentner (probably sent by Walther Bothe), Wilhelm Hanle, Gerhard Hoffmann, and Georg Joos; Peter Debye was invited, but he did not attend.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, CHEMICAL_ELEMENT, SCIENTIST\n\nText:\nThe group included the physicists Walther Bothe, Robert Döpel, Hans Geiger, Wolfgang Gentner (probably sent by Walther Bothe), Wilhelm Hanle, Gerhard Hoffmann, and Georg Joos; Peter Debye was invited, but he did not attend.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_48", "text": "Writing for The Mathematical Gazette on the first edition, L. Rosenhead congratulated Goldstein for a lucid account of classical mechanics leading to modern theoretical physics, which he believed would stand the test of time alongside acknowledged classics such as E. T. Whittaker 's Analytical Dynamics and Arnold Sommerfeld' s Lectures on Theoretical Physics.", "requested_types": ["DISCIPLINE", "ACADEMIC_JOURNAL", "THEORY", "SCIENTIST", "CHEMICAL_ELEMENT", "MISC"], "distractors": ["MISC", "CHEMICAL_ELEMENT", "THEORY"], "gold": {"DISCIPLINE": ["classical mechanics", "theoretical physics", "Analytical Dynamics", "Theoretical Physics"], "ACADEMIC_JOURNAL": ["The Mathematical Gazette"], "THEORY": [], "SCIENTIST": ["L. Rosenhead", "Goldstein", "E. T. Whittaker", "Arnold Sommerfeld"], "CHEMICAL_ELEMENT": [], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ACADEMIC_JOURNAL, THEORY, SCIENTIST, CHEMICAL_ELEMENT, MISC\n\nText:\nWriting for The Mathematical Gazette on the first edition, L. Rosenhead congratulated Goldstein for a lucid account of classical mechanics leading to modern theoretical physics, which he believed would stand the test of time alongside acknowledged classics such as E. T. Whittaker 's Analytical Dynamics and Arnold Sommerfeld' s Lectures on Theoretical Physics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ACADEMIC_JOURNAL, THEORY, SCIENTIST, CHEMICAL_ELEMENT, MISC\n\nText:\nWriting for The Mathematical Gazette on the first edition, L. Rosenhead congratulated Goldstein for a lucid account of classical mechanics leading to modern theoretical physics, which he believed would stand the test of time alongside acknowledged classics such as E. T. Whittaker 's Analytical Dynamics and Arnold Sommerfeld' s Lectures on Theoretical Physics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_49", "text": "During his academic career he served on the editorial boards of the Journal of Rational Mechanics and Analysis, Archive for Rational Mechanics and Analysis, Journal of Elasticity, and the International Journal of Solids and Structures.", "requested_types": ["SCIENTIST", "ACADEMIC_JOURNAL", "MISC"], "distractors": ["SCIENTIST", "MISC"], "gold": {"SCIENTIST": [], "ACADEMIC_JOURNAL": ["Journal of Rational Mechanics and Analysis", "Archive for Rational Mechanics and Analysis", "Journal of Elasticity", "International Journal of Solids and Structures"], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ACADEMIC_JOURNAL, MISC\n\nText:\nDuring his academic career he served on the editorial boards of the Journal of Rational Mechanics and Analysis, Archive for Rational Mechanics and Analysis, Journal of Elasticity, and the International Journal of Solids and Structures.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ACADEMIC_JOURNAL, MISC\n\nText:\nDuring his academic career he served on the editorial boards of the Journal of Rational Mechanics and Analysis, Archive for Rational Mechanics and Analysis, Journal of Elasticity, and the International Journal of Solids and Structures.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_50", "text": "DNA methyltransferase is the enzyme involved in the maintenance of DNA methylation marks.", "requested_types": ["ENZYME", "MISC", "THEORY"], "distractors": ["THEORY"], "gold": {"ENZYME": ["DNA methyltransferase"], "MISC": ["DNA methylation"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, MISC, THEORY\n\nText:\nDNA methyltransferase is the enzyme involved in the maintenance of DNA methylation marks.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, MISC, THEORY\n\nText:\nDNA methyltransferase is the enzyme involved in the maintenance of DNA methylation marks.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_51", "text": "The California portion of the desert also contains Edwards Air Force Base and Naval Air Weapons Station China Lake, noted for experimental aviation and weapons projects, and the Marine Corps Air Ground Combat Center Twentynine Palms in the world at Twentynine Palms.", "requested_types": ["LOCATION", "AWARD", "ASTRONOMICAL_OBJECT"], "distractors": ["ASTRONOMICAL_OBJECT", "AWARD"], "gold": {"LOCATION": ["California", "Edwards Air Force Base", "Naval Air Weapons Station China Lake", "Marine Corps Air Ground Combat Center Twentynine Palms", "Twentynine Palms"], "AWARD": [], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, AWARD, ASTRONOMICAL_OBJECT\n\nText:\nThe California portion of the desert also contains Edwards Air Force Base and Naval Air Weapons Station China Lake, noted for experimental aviation and weapons projects, and the Marine Corps Air Ground Combat Center Twentynine Palms in the world at Twentynine Palms.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, AWARD, ASTRONOMICAL_OBJECT\n\nText:\nThe California portion of the desert also contains Edwards Air Force Base and Naval Air Weapons Station China Lake, noted for experimental aviation and weapons projects, and the Marine Corps Air Ground Combat Center Twentynine Palms in the world at Twentynine Palms.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_52", "text": "Theoretical modelling of two of these super-Earths, Kepler-62e and Kepler-62f, suggests both could be solid, either rocky or rocky with frozen water.", "requested_types": ["ASTRONOMICAL_OBJECT", "MISC", "COUNTRY", "ACADEMIC_JOURNAL"], "distractors": ["COUNTRY", "ACADEMIC_JOURNAL"], "gold": {"ASTRONOMICAL_OBJECT": ["Kepler-62e", "Kepler-62f"], "MISC": ["super-Earths"], "COUNTRY": [], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, COUNTRY, ACADEMIC_JOURNAL\n\nText:\nTheoretical modelling of two of these super-Earths, Kepler-62e and Kepler-62f, suggests both could be solid, either rocky or rocky with frozen water.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, COUNTRY, ACADEMIC_JOURNAL\n\nText:\nTheoretical modelling of two of these super-Earths, Kepler-62e and Kepler-62f, suggests both could be solid, either rocky or rocky with frozen water.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_53", "text": "The group included the physicists Walther Bothe, Robert Döpel, Hans Geiger, Wolfgang Gentner (probably sent by Walther Bothe), Wilhelm Hanle, Gerhard Hoffmann, and Georg Joos; Peter Debye was invited, but he did not attend.", "requested_types": ["PROTEIN", "SCIENTIST"], "distractors": ["PROTEIN"], "gold": {"PROTEIN": [], "SCIENTIST": ["Walther Bothe", "Robert Döpel", "Hans Geiger", "Wolfgang Gentner", "Wilhelm Hanle", "Gerhard Hoffmann", "Georg Joos", "Peter Debye"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST\n\nText:\nThe group included the physicists Walther Bothe, Robert Döpel, Hans Geiger, Wolfgang Gentner (probably sent by Walther Bothe), Wilhelm Hanle, Gerhard Hoffmann, and Georg Joos; Peter Debye was invited, but he did not attend.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST\n\nText:\nThe group included the physicists Walther Bothe, Robert Döpel, Hans Geiger, Wolfgang Gentner (probably sent by Walther Bothe), Wilhelm Hanle, Gerhard Hoffmann, and Georg Joos; Peter Debye was invited, but he did not attend.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_54", "text": "Nature Research also publishes other specialized journals including Nature Neuroscience, Nature Biotechnology, Nature Methods, the Nature Clinical Practice series of journals, Nature Structural & Molecular Biology, Nature Chemistry, and the Nature Reviews series of journals.", "requested_types": ["ORGANISATION", "MISC", "SCIENTIST", "ACADEMIC_JOURNAL"], "distractors": ["SCIENTIST"], "gold": {"ORGANISATION": ["Nature Research"], "MISC": ["Nature Clinical Practice", "Nature Reviews"], "SCIENTIST": [], "ACADEMIC_JOURNAL": ["Nature Neuroscience", "Nature Biotechnology", "Nature Methods", "Nature Structural & Molecular Biology", "Nature Chemistry"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, SCIENTIST, ACADEMIC_JOURNAL\n\nText:\nNature Research also publishes other specialized journals including Nature Neuroscience, Nature Biotechnology, Nature Methods, the Nature Clinical Practice series of journals, Nature Structural & Molecular Biology, Nature Chemistry, and the Nature Reviews series of journals.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, SCIENTIST, ACADEMIC_JOURNAL\n\nText:\nNature Research also publishes other specialized journals including Nature Neuroscience, Nature Biotechnology, Nature Methods, the Nature Clinical Practice series of journals, Nature Structural & Molecular Biology, Nature Chemistry, and the Nature Reviews series of journals.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_55", "text": "Biju has extensively published in high impact factor and prestigious journals such as Nature, Science, Zoological Journal of the Linnean Society, Proceedings of the National Academy of Sciences, Proceedings of the Royal Society, and has written books on both plants and amphibians.", "requested_types": ["PROTEIN", "ENZYME", "SCIENTIST", "ACADEMIC_JOURNAL"], "distractors": ["PROTEIN", "ENZYME"], "gold": {"PROTEIN": [], "ENZYME": [], "SCIENTIST": ["Biju"], "ACADEMIC_JOURNAL": ["Nature", "Science", "Zoological Journal of the Linnean Society", "Proceedings of the National Academy of Sciences", "Proceedings of the Royal Society"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ENZYME, SCIENTIST, ACADEMIC_JOURNAL\n\nText:\nBiju has extensively published in high impact factor and prestigious journals such as Nature, Science, Zoological Journal of the Linnean Society, Proceedings of the National Academy of Sciences, Proceedings of the Royal Society, and has written books on both plants and amphibians.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ENZYME, SCIENTIST, ACADEMIC_JOURNAL\n\nText:\nBiju has extensively published in high impact factor and prestigious journals such as Nature, Science, Zoological Journal of the Linnean Society, Proceedings of the National Academy of Sciences, Proceedings of the Royal Society, and has written books on both plants and amphibians.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_56", "text": "Lexell calculated its orbit, showed that the comet had had a much larger perihelion before the encounter with Jupiter in 1767 and predicted that after encountering Jupiter again in 1779 it would be altogether expelled from the inner Solar System.", "requested_types": ["SCIENTIST", "PERSON", "ASTRONOMICAL_OBJECT", "COUNTRY", "MISC"], "distractors": ["PERSON", "COUNTRY"], "gold": {"SCIENTIST": ["Lexell"], "PERSON": [], "ASTRONOMICAL_OBJECT": ["Jupiter"], "COUNTRY": [], "MISC": ["Solar System"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, PERSON, ASTRONOMICAL_OBJECT, COUNTRY, MISC\n\nText:\nLexell calculated its orbit, showed that the comet had had a much larger perihelion before the encounter with Jupiter in 1767 and predicted that after encountering Jupiter again in 1779 it would be altogether expelled from the inner Solar System.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, PERSON, ASTRONOMICAL_OBJECT, COUNTRY, MISC\n\nText:\nLexell calculated its orbit, showed that the comet had had a much larger perihelion before the encounter with Jupiter in 1767 and predicted that after encountering Jupiter again in 1779 it would be altogether expelled from the inner Solar System.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_57", "text": "This minor planet was named in memory of Russian plant geneticist Nikolai Vavilov (1887-1943) and his physicist brother Sergey Ivanovich Vavilov (1891-1951).", "requested_types": ["MISC", "SCIENTIST", "DISCIPLINE", "ORGANISATION", "UNIVERSITY"], "distractors": ["DISCIPLINE", "ORGANISATION", "UNIVERSITY"], "gold": {"MISC": ["minor planet", "Russian"], "SCIENTIST": ["Nikolai Vavilov", "Sergey Ivanovich Vavilov"], "DISCIPLINE": [], "ORGANISATION": [], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST, DISCIPLINE, ORGANISATION, UNIVERSITY\n\nText:\nThis minor planet was named in memory of Russian plant geneticist Nikolai Vavilov (1887-1943) and his physicist brother Sergey Ivanovich Vavilov (1891-1951).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST, DISCIPLINE, ORGANISATION, UNIVERSITY\n\nText:\nThis minor planet was named in memory of Russian plant geneticist Nikolai Vavilov (1887-1943) and his physicist brother Sergey Ivanovich Vavilov (1891-1951).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_58", "text": "Some of the many prestigious residents include: Academy Award winners Rustam Ibrahimbeyov and Vladimir Menshov, one of the founders and head of the Soviet space program Kerim Kerimov, Nobel Prize winner and physicist Lev Landau and famous musicians such as Gara Garayev, Uzeyir Hajibeyov, Muslim Magomayev, Vagif Mustafazadeh and Alim Qasimov.", "requested_types": ["ORGANISATION", "LOCATION", "PERSON", "AWARD", "ASTRONOMICAL_OBJECT", "SCIENTIST"], "distractors": ["ASTRONOMICAL_OBJECT", "LOCATION"], "gold": {"ORGANISATION": ["Soviet space program"], "LOCATION": [], "PERSON": ["Rustam Ibrahimbeyov", "Vladimir Menshov", "Gara Garayev", "Uzeyir Hajibeyov", "Muslim Magomayev", "Vagif Mustafazadeh", "Alim Qasimov"], "AWARD": ["Academy Award", "Nobel Prize"], "ASTRONOMICAL_OBJECT": [], "SCIENTIST": ["Kerim Kerimov", "Lev Landau"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, LOCATION, PERSON, AWARD, ASTRONOMICAL_OBJECT, SCIENTIST\n\nText:\nSome of the many prestigious residents include: Academy Award winners Rustam Ibrahimbeyov and Vladimir Menshov, one of the founders and head of the Soviet space program Kerim Kerimov, Nobel Prize winner and physicist Lev Landau and famous musicians such as Gara Garayev, Uzeyir Hajibeyov, Muslim Magomayev, Vagif Mustafazadeh and Alim Qasimov.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, LOCATION, PERSON, AWARD, ASTRONOMICAL_OBJECT, SCIENTIST\n\nText:\nSome of the many prestigious residents include: Academy Award winners Rustam Ibrahimbeyov and Vladimir Menshov, one of the founders and head of the Soviet space program Kerim Kerimov, Nobel Prize winner and physicist Lev Landau and famous musicians such as Gara Garayev, Uzeyir Hajibeyov, Muslim Magomayev, Vagif Mustafazadeh and Alim Qasimov.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_59", "text": "From 1937 to 1942, as successor to Max Delbrück, Flügge was an assistant to Otto Hahn at the Kaiser-Wilhelm Institut für Chemie (KWIC, after World War II reorganized and renamed the Max Planck Institute for Chemistry), in Berlin-Dahlem.", "requested_types": ["LOCATION", "ORGANISATION", "SCIENTIST", "ENZYME", "EVENT"], "distractors": ["ENZYME"], "gold": {"LOCATION": ["Berlin-Dahlem"], "ORGANISATION": ["Kaiser-Wilhelm Institut für Chemie", "KWIC", "Max Planck Institute for Chemistry"], "SCIENTIST": ["Max Delbrück", "Flügge", "Otto Hahn"], "ENZYME": [], "EVENT": ["World War II"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ORGANISATION, SCIENTIST, ENZYME, EVENT\n\nText:\nFrom 1937 to 1942, as successor to Max Delbrück, Flügge was an assistant to Otto Hahn at the Kaiser-Wilhelm Institut für Chemie (KWIC, after World War II reorganized and renamed the Max Planck Institute for Chemistry), in Berlin-Dahlem.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ORGANISATION, SCIENTIST, ENZYME, EVENT\n\nText:\nFrom 1937 to 1942, as successor to Max Delbrück, Flügge was an assistant to Otto Hahn at the Kaiser-Wilhelm Institut für Chemie (KWIC, after World War II reorganized and renamed the Max Planck Institute for Chemistry), in Berlin-Dahlem.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_60", "text": "In 2007 the British Biochemical Society was given a grant by the Wellcome Trust to catalogue and preserve the 35 laboratory notebooks in which Sanger recorded his research from 1944 to 1983.", "requested_types": ["PROTEIN", "SCIENTIST", "EVENT", "ORGANISATION"], "distractors": ["EVENT", "PROTEIN"], "gold": {"PROTEIN": [], "SCIENTIST": ["Sanger"], "EVENT": [], "ORGANISATION": ["British Biochemical Society", "Wellcome Trust"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST, EVENT, ORGANISATION\n\nText:\nIn 2007 the British Biochemical Society was given a grant by the Wellcome Trust to catalogue and preserve the 35 laboratory notebooks in which Sanger recorded his research from 1944 to 1983.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST, EVENT, ORGANISATION\n\nText:\nIn 2007 the British Biochemical Society was given a grant by the Wellcome Trust to catalogue and preserve the 35 laboratory notebooks in which Sanger recorded his research from 1944 to 1983.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_61", "text": "He won the J. Lawrence Smith Medal from the National Academy of Sciences in 1960, the Meteoritical Society Frederick C. Leonard Memorial Medal in 1968, the Kepler Gold Medal from the American Association for the Advancement of Science & amp; Meteoritical Society in 1972, the Gold Medal of the Royal Astronomical Society in 1975 and the Bruce Medal in 1976.", "requested_types": ["AWARD", "ORGANISATION", "THEORY", "DISCIPLINE", "EVENT"], "distractors": ["DISCIPLINE", "THEORY", "EVENT"], "gold": {"AWARD": ["J. Lawrence Smith Medal", "Meteoritical Society Frederick C. Leonard Memorial Medal", "Kepler Gold Medal", "Gold Medal of the Royal Astronomical Society", "Bruce Medal"], "ORGANISATION": ["National Academy of Sciences", "American Association"], "THEORY": [], "DISCIPLINE": [], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ORGANISATION, THEORY, DISCIPLINE, EVENT\n\nText:\nHe won the J. Lawrence Smith Medal from the National Academy of Sciences in 1960, the Meteoritical Society Frederick C. Leonard Memorial Medal in 1968, the Kepler Gold Medal from the American Association for the Advancement of Science & amp; Meteoritical Society in 1972, the Gold Medal of the Royal Astronomical Society in 1975 and the Bruce Medal in 1976.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ORGANISATION, THEORY, DISCIPLINE, EVENT\n\nText:\nHe won the J. Lawrence Smith Medal from the National Academy of Sciences in 1960, the Meteoritical Society Frederick C. Leonard Memorial Medal in 1968, the Kepler Gold Medal from the American Association for the Advancement of Science & amp; Meteoritical Society in 1972, the Gold Medal of the Royal Astronomical Society in 1975 and the Bruce Medal in 1976.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_62", "text": "On August 11, 2014, astronomers released studies, using the Atacama Large Millimeter / Submillimeter Array (ALMA) for the first time, that detailed the distribution of Hydrogen cyanide, Hydrogen isocyanide, Formaldehyde, and dust inside the comae of comet s C / 2012 F6 (Lemmon) and C / 2012 S1 (ISON).", "requested_types": ["ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND", "MISC", "AWARD", "ENZYME"], "distractors": ["AWARD", "ENZYME"], "gold": {"ASTRONOMICAL_OBJECT": ["comet s C / 2012 F6", "Lemmon", "C / 2012 S1", "ISON"], "CHEMICAL_COMPOUND": ["Hydrogen cyanide", "Hydrogen isocyanide", "Formaldehyde"], "MISC": ["Atacama Large Millimeter / Submillimeter Array", "ALMA"], "AWARD": [], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, MISC, AWARD, ENZYME\n\nText:\nOn August 11, 2014, astronomers released studies, using the Atacama Large Millimeter / Submillimeter Array (ALMA) for the first time, that detailed the distribution of Hydrogen cyanide, Hydrogen isocyanide, Formaldehyde, and dust inside the comae of comet s C / 2012 F6 (Lemmon) and C / 2012 S1 (ISON).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, MISC, AWARD, ENZYME\n\nText:\nOn August 11, 2014, astronomers released studies, using the Atacama Large Millimeter / Submillimeter Array (ALMA) for the first time, that detailed the distribution of Hydrogen cyanide, Hydrogen isocyanide, Formaldehyde, and dust inside the comae of comet s C / 2012 F6 (Lemmon) and C / 2012 S1 (ISON).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_63", "text": "Ran exists in the cell in two nucleotide-bound forms: Guanosine diphosphate -bound and Guanosine triphosphate -bound.", "requested_types": ["PROTEIN", "CHEMICAL_COMPOUND", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"PROTEIN": ["Ran"], "CHEMICAL_COMPOUND": ["Guanosine diphosphate", "Guanosine triphosphate"], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_COMPOUND, CHEMICAL_ELEMENT\n\nText:\nRan exists in the cell in two nucleotide-bound forms: Guanosine diphosphate -bound and Guanosine triphosphate -bound.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_COMPOUND, CHEMICAL_ELEMENT\n\nText:\nRan exists in the cell in two nucleotide-bound forms: Guanosine diphosphate -bound and Guanosine triphosphate -bound.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_64", "text": "The WH2 domain is found as a modular part of larger proteins; it can be associated with the WH1 domain or EVH1 domain and with the CRIB domain, and the WH2 domain can occur as a tandem repeat.", "requested_types": ["MISC", "CHEMICAL_COMPOUND", "ACADEMIC_JOURNAL"], "distractors": ["CHEMICAL_COMPOUND", "ACADEMIC_JOURNAL"], "gold": {"MISC": ["WH2 domain", "WH1 domain", "EVH1 domain", "CRIB domain"], "CHEMICAL_COMPOUND": [], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_COMPOUND, ACADEMIC_JOURNAL\n\nText:\nThe WH2 domain is found as a modular part of larger proteins; it can be associated with the WH1 domain or EVH1 domain and with the CRIB domain, and the WH2 domain can occur as a tandem repeat.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_COMPOUND, ACADEMIC_JOURNAL\n\nText:\nThe WH2 domain is found as a modular part of larger proteins; it can be associated with the WH1 domain or EVH1 domain and with the CRIB domain, and the WH2 domain can occur as a tandem repeat.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_65", "text": "The song was also nominated for a Golden Globe Award for Best Original Song and an Academy Award for Best Original Song.", "requested_types": ["AWARD", "PERSON", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_ELEMENT", "PERSON"], "gold": {"AWARD": ["Golden Globe Award for Best Original Song", "Academy Award for Best Original Song"], "PERSON": [], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, PERSON, CHEMICAL_ELEMENT\n\nText:\nThe song was also nominated for a Golden Globe Award for Best Original Song and an Academy Award for Best Original Song.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, PERSON, CHEMICAL_ELEMENT\n\nText:\nThe song was also nominated for a Golden Globe Award for Best Original Song and an Academy Award for Best Original Song.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_66", "text": "A United States Air Force KC-135 Stratotanker arrived at Misawa Air Base on 14 March with the first batch of relief workers and 50 civil engineers from Kadena Air Base.", "requested_types": ["DISCIPLINE", "PERSON", "LOCATION", "ENZYME", "MISC"], "distractors": ["DISCIPLINE", "PERSON", "ENZYME"], "gold": {"DISCIPLINE": [], "PERSON": [], "LOCATION": ["Misawa Air Base", "Kadena Air Base"], "ENZYME": [], "MISC": ["United States Air Force KC-135 Stratotanker"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, PERSON, LOCATION, ENZYME, MISC\n\nText:\nA United States Air Force KC-135 Stratotanker arrived at Misawa Air Base on 14 March with the first batch of relief workers and 50 civil engineers from Kadena Air Base.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, PERSON, LOCATION, ENZYME, MISC\n\nText:\nA United States Air Force KC-135 Stratotanker arrived at Misawa Air Base on 14 March with the first batch of relief workers and 50 civil engineers from Kadena Air Base.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_67", "text": "The journal is abstracted and indexed in Applied Mechanics Reviews, Current Contents / Engineering, Computing & Technology, Current Contents / Physics, Chemical,; Earth Sciences, Compendex, Inspec, Mathematical Reviews, Scopus, and Zentralblatt MATH.", "requested_types": ["MISC", "SCIENTIST", "ACADEMIC_JOURNAL", "ENZYME", "DISCIPLINE"], "distractors": ["SCIENTIST", "ENZYME"], "gold": {"MISC": ["Compendex", "Inspec", "Scopus", "Zentralblatt MATH"], "SCIENTIST": [], "ACADEMIC_JOURNAL": ["Applied Mechanics Reviews", "Mathematical Reviews"], "ENZYME": [], "DISCIPLINE": ["Engineering", "Computing & Technology", "Physics", "Chemical", "Earth Sciences"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST, ACADEMIC_JOURNAL, ENZYME, DISCIPLINE\n\nText:\nThe journal is abstracted and indexed in Applied Mechanics Reviews, Current Contents / Engineering, Computing & Technology, Current Contents / Physics, Chemical,; Earth Sciences, Compendex, Inspec, Mathematical Reviews, Scopus, and Zentralblatt MATH.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST, ACADEMIC_JOURNAL, ENZYME, DISCIPLINE\n\nText:\nThe journal is abstracted and indexed in Applied Mechanics Reviews, Current Contents / Engineering, Computing & Technology, Current Contents / Physics, Chemical,; Earth Sciences, Compendex, Inspec, Mathematical Reviews, Scopus, and Zentralblatt MATH.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_68", "text": "Neurexin - Neuroligin interactions help establish the trans-synaptic functional asymmetry essential for the stabilization and maintenance of proper synaptic transmission.", "requested_types": ["PROTEIN", "CHEMICAL_COMPOUND", "MISC"], "distractors": ["CHEMICAL_COMPOUND"], "gold": {"PROTEIN": ["Neurexin", "Neuroligin"], "CHEMICAL_COMPOUND": [], "MISC": ["synaptic transmission"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_COMPOUND, MISC\n\nText:\nNeurexin - Neuroligin interactions help establish the trans-synaptic functional asymmetry essential for the stabilization and maintenance of proper synaptic transmission.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_COMPOUND, MISC\n\nText:\nNeurexin - Neuroligin interactions help establish the trans-synaptic functional asymmetry essential for the stabilization and maintenance of proper synaptic transmission.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_69", "text": "In the 1950s, James D. Watson, Francis Crick, Rosalind Franklin, and Maurice Wilkins were instrumental in solving DNA structure and suggesting its relationship with the genetic transfer of information.", "requested_types": ["ASTRONOMICAL_OBJECT", "SCIENTIST", "PERSON", "MISC", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND", "PERSON", "ASTRONOMICAL_OBJECT"], "gold": {"ASTRONOMICAL_OBJECT": [], "SCIENTIST": ["James D. Watson", "Francis Crick", "Rosalind Franklin", "Maurice Wilkins"], "PERSON": [], "MISC": ["DNA"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, SCIENTIST, PERSON, MISC, CHEMICAL_COMPOUND\n\nText:\nIn the 1950s, James D. Watson, Francis Crick, Rosalind Franklin, and Maurice Wilkins were instrumental in solving DNA structure and suggesting its relationship with the genetic transfer of information.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, SCIENTIST, PERSON, MISC, CHEMICAL_COMPOUND\n\nText:\nIn the 1950s, James D. Watson, Francis Crick, Rosalind Franklin, and Maurice Wilkins were instrumental in solving DNA structure and suggesting its relationship with the genetic transfer of information.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_70", "text": "In July 1989, an Indian group from the Bhabha Atomic Research Centre (P. K. Iyengar and M. Srinivasan) and in October 1989, John Bockris' group from Texas University reported on the creation of tritium.", "requested_types": ["PROTEIN", "CHEMICAL_ELEMENT", "ORGANISATION", "UNIVERSITY", "SCIENTIST", "MISC"], "distractors": ["PROTEIN"], "gold": {"PROTEIN": [], "CHEMICAL_ELEMENT": ["tritium"], "ORGANISATION": ["Bhabha Atomic Research Centre"], "UNIVERSITY": ["Texas University"], "SCIENTIST": ["P. K. Iyengar", "M. Srinivasan", "John Bockris"], "MISC": ["Indian"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_ELEMENT, ORGANISATION, UNIVERSITY, SCIENTIST, MISC\n\nText:\nIn July 1989, an Indian group from the Bhabha Atomic Research Centre (P. K. Iyengar and M. Srinivasan) and in October 1989, John Bockris' group from Texas University reported on the creation of tritium.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_ELEMENT, ORGANISATION, UNIVERSITY, SCIENTIST, MISC\n\nText:\nIn July 1989, an Indian group from the Bhabha Atomic Research Centre (P. K. Iyengar and M. Srinivasan) and in October 1989, John Bockris' group from Texas University reported on the creation of tritium.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_71", "text": "Ring1A and Ring1B (which are also known as Rnf1 and Rnf2, respectively) are Ubiquitin ligase that mark lysine 119 of histone H2A with a single ubiquitin group (H2AK119ub).", "requested_types": ["ENZYME", "MISC", "COUNTRY", "PROTEIN", "SCIENTIST"], "distractors": ["SCIENTIST", "COUNTRY"], "gold": {"ENZYME": ["Ubiquitin ligase"], "MISC": ["lysine 119", "H2A", "H2AK119ub"], "COUNTRY": [], "PROTEIN": ["Ring1A", "Ring1B", "Rnf1", "Rnf2"], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, MISC, COUNTRY, PROTEIN, SCIENTIST\n\nText:\nRing1A and Ring1B (which are also known as Rnf1 and Rnf2, respectively) are Ubiquitin ligase that mark lysine 119 of histone H2A with a single ubiquitin group (H2AK119ub).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, MISC, COUNTRY, PROTEIN, SCIENTIST\n\nText:\nRing1A and Ring1B (which are also known as Rnf1 and Rnf2, respectively) are Ubiquitin ligase that mark lysine 119 of histone H2A with a single ubiquitin group (H2AK119ub).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_72", "text": "The concept had been advocated by Jean-Baptiste Lamarck, Étienne Geoffroy Saint-Hilaire, Erasmus Darwin, and Robert Grant, among others.", "requested_types": ["SCIENTIST", "PERSON"], "distractors": ["PERSON"], "gold": {"SCIENTIST": ["Jean-Baptiste Lamarck", "Étienne Geoffroy Saint-Hilaire", "Erasmus Darwin", "Robert Grant"], "PERSON": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, PERSON\n\nText:\nThe concept had been advocated by Jean-Baptiste Lamarck, Étienne Geoffroy Saint-Hilaire, Erasmus Darwin, and Robert Grant, among others.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, PERSON\n\nText:\nThe concept had been advocated by Jean-Baptiste Lamarck, Étienne Geoffroy Saint-Hilaire, Erasmus Darwin, and Robert Grant, among others.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_73", "text": "On 11 August 2014, astronomers released studies, using the Atacama Large Millimeter / Submillimeter Array (ALMA) for the first time, that detailed the distribution of HCN, Hydrogen isocyanide, Formaldehyde, and dust inside the comae of comet s.", "requested_types": ["ENZYME", "MISC", "ACADEMIC_JOURNAL", "CHEMICAL_COMPOUND", "LOCATION"], "distractors": ["ACADEMIC_JOURNAL", "LOCATION", "ENZYME"], "gold": {"ENZYME": [], "MISC": ["Atacama Large Millimeter / Submillimeter Array", "ALMA"], "ACADEMIC_JOURNAL": [], "CHEMICAL_COMPOUND": ["Hydrogen isocyanide", "Formaldehyde"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, MISC, ACADEMIC_JOURNAL, CHEMICAL_COMPOUND, LOCATION\n\nText:\nOn 11 August 2014, astronomers released studies, using the Atacama Large Millimeter / Submillimeter Array (ALMA) for the first time, that detailed the distribution of HCN, Hydrogen isocyanide, Formaldehyde, and dust inside the comae of comet s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, MISC, ACADEMIC_JOURNAL, CHEMICAL_COMPOUND, LOCATION\n\nText:\nOn 11 August 2014, astronomers released studies, using the Atacama Large Millimeter / Submillimeter Array (ALMA) for the first time, that detailed the distribution of HCN, Hydrogen isocyanide, Formaldehyde, and dust inside the comae of comet s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_74", "text": "He is currently on the editorial board of more than 10 scientific journals, among them Chaos, Philosophical Transactions of the Royal Society A, PLOS One, European Journal of Physics, J. Nonlinear Science and Nonlinear Processes in Geophysics and of the Springer Series Complexity.", "requested_types": ["COUNTRY", "ORGANISATION", "ACADEMIC_JOURNAL"], "distractors": ["ORGANISATION", "COUNTRY"], "gold": {"COUNTRY": [], "ORGANISATION": [], "ACADEMIC_JOURNAL": ["Chaos", "Philosophical Transactions of the Royal Society A", "PLOS One", "European Journal of Physics", "J. Nonlinear Science", "Nonlinear Processes in Geophysics", "the Springer Series Complexity"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ORGANISATION, ACADEMIC_JOURNAL\n\nText:\nHe is currently on the editorial board of more than 10 scientific journals, among them Chaos, Philosophical Transactions of the Royal Society A, PLOS One, European Journal of Physics, J. Nonlinear Science and Nonlinear Processes in Geophysics and of the Springer Series Complexity.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ORGANISATION, ACADEMIC_JOURNAL\n\nText:\nHe is currently on the editorial board of more than 10 scientific journals, among them Chaos, Philosophical Transactions of the Royal Society A, PLOS One, European Journal of Physics, J. Nonlinear Science and Nonlinear Processes in Geophysics and of the Springer Series Complexity.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_75", "text": "This unification, which was observed by Michael Faraday, extended by James Clerk Maxwell, and partially reformulated by Oliver Heaviside and Heinrich Hertz, is one of the key accomplishments of 19th-century mathematical physics.", "requested_types": ["UNIVERSITY", "CHEMICAL_ELEMENT", "DISCIPLINE", "SCIENTIST", "AWARD"], "distractors": ["CHEMICAL_ELEMENT", "UNIVERSITY", "AWARD"], "gold": {"UNIVERSITY": [], "CHEMICAL_ELEMENT": [], "DISCIPLINE": ["mathematical physics"], "SCIENTIST": ["Michael Faraday", "James Clerk Maxwell", "Oliver Heaviside", "Heinrich Hertz"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, CHEMICAL_ELEMENT, DISCIPLINE, SCIENTIST, AWARD\n\nText:\nThis unification, which was observed by Michael Faraday, extended by James Clerk Maxwell, and partially reformulated by Oliver Heaviside and Heinrich Hertz, is one of the key accomplishments of 19th-century mathematical physics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, CHEMICAL_ELEMENT, DISCIPLINE, SCIENTIST, AWARD\n\nText:\nThis unification, which was observed by Michael Faraday, extended by James Clerk Maxwell, and partially reformulated by Oliver Heaviside and Heinrich Hertz, is one of the key accomplishments of 19th-century mathematical physics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_76", "text": "He has co-authored a number of frequently-cited articles in Circulation Research, the Journal of Biological Chemistry, and the American Journal of Physiology.", "requested_types": ["ACADEMIC_JOURNAL", "SCIENTIST", "ENZYME"], "distractors": ["ENZYME", "SCIENTIST"], "gold": {"ACADEMIC_JOURNAL": ["Circulation Research", "Journal of Biological Chemistry", "American Journal of Physiology"], "SCIENTIST": [], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, ENZYME\n\nText:\nHe has co-authored a number of frequently-cited articles in Circulation Research, the Journal of Biological Chemistry, and the American Journal of Physiology.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, ENZYME\n\nText:\nHe has co-authored a number of frequently-cited articles in Circulation Research, the Journal of Biological Chemistry, and the American Journal of Physiology.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_77", "text": "Historian J. G. M. Ramsey reported a conference between militia commander John Sevier and Cherokee Chief Hanging Maw held at the original Citico in 1782 in which the two sides agreed to a truce.", "requested_types": ["ENZYME", "LOCATION", "PERSON", "ORGANISATION", "EVENT", "MISC", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_ELEMENT", "EVENT", "ENZYME"], "gold": {"ENZYME": [], "LOCATION": ["Citico"], "PERSON": ["J. G. M. Ramsey", "John Sevier", "Hanging Maw"], "ORGANISATION": ["militia"], "EVENT": [], "MISC": ["Cherokee"], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, LOCATION, PERSON, ORGANISATION, EVENT, MISC, CHEMICAL_ELEMENT\n\nText:\nHistorian J. G. M. Ramsey reported a conference between militia commander John Sevier and Cherokee Chief Hanging Maw held at the original Citico in 1782 in which the two sides agreed to a truce.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, LOCATION, PERSON, ORGANISATION, EVENT, MISC, CHEMICAL_ELEMENT\n\nText:\nHistorian J. G. M. Ramsey reported a conference between militia commander John Sevier and Cherokee Chief Hanging Maw held at the original Citico in 1782 in which the two sides agreed to a truce.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_78", "text": "Young began to study medicine in London at St Bartholomew's Hospital in 1792, moved to the University of Edinburgh Medical School in 1794, and a year later went to Göttingen, Lower Saxony, Germany, where he obtained the degree of doctor of medicine in 1796 from the University of Göttingen.", "requested_types": ["COUNTRY", "LOCATION", "UNIVERSITY", "ORGANISATION", "AWARD"], "distractors": ["AWARD"], "gold": {"COUNTRY": ["Germany"], "LOCATION": ["London", "Göttingen", "Lower Saxony"], "UNIVERSITY": ["University of Edinburgh Medical School", "University of Göttingen"], "ORGANISATION": ["St Bartholomew's Hospital"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, LOCATION, UNIVERSITY, ORGANISATION, AWARD\n\nText:\nYoung began to study medicine in London at St Bartholomew's Hospital in 1792, moved to the University of Edinburgh Medical School in 1794, and a year later went to Göttingen, Lower Saxony, Germany, where he obtained the degree of doctor of medicine in 1796 from the University of Göttingen.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, LOCATION, UNIVERSITY, ORGANISATION, AWARD\n\nText:\nYoung began to study medicine in London at St Bartholomew's Hospital in 1792, moved to the University of Edinburgh Medical School in 1794, and a year later went to Göttingen, Lower Saxony, Germany, where he obtained the degree of doctor of medicine in 1796 from the University of Göttingen.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_79", "text": "In 1827 he was elected member of the Royal Swedish Academy of Sciences, and in 1829 he won the Gold Medal of the Royal Astronomical Society.", "requested_types": ["ORGANISATION", "SCIENTIST", "AWARD"], "distractors": ["SCIENTIST"], "gold": {"ORGANISATION": ["Royal Swedish Academy of Sciences"], "SCIENTIST": [], "AWARD": ["Gold Medal of the Royal Astronomical Society"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, SCIENTIST, AWARD\n\nText:\nIn 1827 he was elected member of the Royal Swedish Academy of Sciences, and in 1829 he won the Gold Medal of the Royal Astronomical Society.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, SCIENTIST, AWARD\n\nText:\nIn 1827 he was elected member of the Royal Swedish Academy of Sciences, and in 1829 he won the Gold Medal of the Royal Astronomical Society.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_80", "text": "The album earned Usher numerous awards, including two Grammy Award s, three Billboard Music Awards and a BET Awards.", "requested_types": ["AWARD", "PROTEIN", "ENZYME", "PERSON", "SCIENTIST"], "distractors": ["PROTEIN", "ENZYME", "SCIENTIST"], "gold": {"AWARD": ["Grammy Award", "Billboard Music Awards", "BET Awards"], "PROTEIN": [], "ENZYME": [], "PERSON": ["Usher"], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, PROTEIN, ENZYME, PERSON, SCIENTIST\n\nText:\nThe album earned Usher numerous awards, including two Grammy Award s, three Billboard Music Awards and a BET Awards.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, PROTEIN, ENZYME, PERSON, SCIENTIST\n\nText:\nThe album earned Usher numerous awards, including two Grammy Award s, three Billboard Music Awards and a BET Awards.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_81", "text": "Amphitrite (minor planet designation: 29 Amphridite) is one of the largest S-type asteroid s, approximately in diameter, and probably third largest after 15 Eunomia and 3 Juno, although 7 Iris and 532 Herculina are similar in size.", "requested_types": ["ASTRONOMICAL_OBJECT", "PERSON", "PROTEIN", "MISC", "CHEMICAL_ELEMENT"], "distractors": ["PROTEIN", "CHEMICAL_ELEMENT", "PERSON"], "gold": {"ASTRONOMICAL_OBJECT": ["Amphitrite", "29 Amphridite", "15 Eunomia", "3 Juno", "7 Iris", "532 Herculina"], "PERSON": [], "PROTEIN": [], "MISC": ["minor planet"], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, PERSON, PROTEIN, MISC, CHEMICAL_ELEMENT\n\nText:\nAmphitrite (minor planet designation: 29 Amphridite) is one of the largest S-type asteroid s, approximately in diameter, and probably third largest after 15 Eunomia and 3 Juno, although 7 Iris and 532 Herculina are similar in size.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, PERSON, PROTEIN, MISC, CHEMICAL_ELEMENT\n\nText:\nAmphitrite (minor planet designation: 29 Amphridite) is one of the largest S-type asteroid s, approximately in diameter, and probably third largest after 15 Eunomia and 3 Juno, although 7 Iris and 532 Herculina are similar in size.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_82", "text": "Woese was a MacArthur Fellow in 1984, was made a member of the National Academy of Sciences in 1988, received the Leeuwenhoek Medal (microbiology's highest honor) in 1992, the Selman A. Waksman Award in Microbiology in 1995 from the National Academy of Sciences,", "requested_types": ["ORGANISATION", "SCIENTIST", "AWARD", "DISCIPLINE", "ACADEMIC_JOURNAL", "LOCATION", "CHEMICAL_COMPOUND"], "distractors": ["ACADEMIC_JOURNAL", "LOCATION", "CHEMICAL_COMPOUND"], "gold": {"ORGANISATION": ["National Academy of Sciences"], "SCIENTIST": ["Woese"], "AWARD": ["MacArthur Fellow", "Leeuwenhoek Medal", "Selman A. Waksman Award in Microbiology"], "DISCIPLINE": ["microbiology"], "ACADEMIC_JOURNAL": [], "LOCATION": [], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, SCIENTIST, AWARD, DISCIPLINE, ACADEMIC_JOURNAL, LOCATION, CHEMICAL_COMPOUND\n\nText:\nWoese was a MacArthur Fellow in 1984, was made a member of the National Academy of Sciences in 1988, received the Leeuwenhoek Medal (microbiology's highest honor) in 1992, the Selman A. Waksman Award in Microbiology in 1995 from the National Academy of Sciences,\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, SCIENTIST, AWARD, DISCIPLINE, ACADEMIC_JOURNAL, LOCATION, CHEMICAL_COMPOUND\n\nText:\nWoese was a MacArthur Fellow in 1984, was made a member of the National Academy of Sciences in 1988, received the Leeuwenhoek Medal (microbiology's highest honor) in 1992, the Selman A. Waksman Award in Microbiology in 1995 from the National Academy of Sciences,\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_83", "text": "She played at 1989 AFC Championship and 1990 Asian Games.", "requested_types": ["ACADEMIC_JOURNAL", "EVENT"], "distractors": ["ACADEMIC_JOURNAL"], "gold": {"ACADEMIC_JOURNAL": [], "EVENT": ["1989 AFC Championship", "1990 Asian Games"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, EVENT\n\nText:\nShe played at 1989 AFC Championship and 1990 Asian Games.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, EVENT\n\nText:\nShe played at 1989 AFC Championship and 1990 Asian Games.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_84", "text": "He competed in Swimming at the 1988 Summer Olympics at the 1988 Summer Olympics.", "requested_types": ["LOCATION", "ORGANISATION", "EVENT", "ENZYME"], "distractors": ["ENZYME", "ORGANISATION", "LOCATION"], "gold": {"LOCATION": [], "ORGANISATION": [], "EVENT": ["1988 Summer Olympics"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ORGANISATION, EVENT, ENZYME\n\nText:\nHe competed in Swimming at the 1988 Summer Olympics at the 1988 Summer Olympics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ORGANISATION, EVENT, ENZYME\n\nText:\nHe competed in Swimming at the 1988 Summer Olympics at the 1988 Summer Olympics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_85", "text": "Binding of these proteins recruit Histone deacetylase enzyme which initiate chromatin remodeling such that the DNA becoming less accessible to transcriptional machinery, such as RNA polymerase, effectively repressing gene expression.", "requested_types": ["MISC", "ENZYME", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND"], "gold": {"MISC": ["chromatin remodeling", "DNA"], "ENZYME": ["Histone deacetylase enzyme", "RNA polymerase"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ENZYME, CHEMICAL_COMPOUND\n\nText:\nBinding of these proteins recruit Histone deacetylase enzyme which initiate chromatin remodeling such that the DNA becoming less accessible to transcriptional machinery, such as RNA polymerase, effectively repressing gene expression.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ENZYME, CHEMICAL_COMPOUND\n\nText:\nBinding of these proteins recruit Histone deacetylase enzyme which initiate chromatin remodeling such that the DNA becoming less accessible to transcriptional machinery, such as RNA polymerase, effectively repressing gene expression.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_86", "text": "Lycomedes was discovered on 26 September 1960, by Dutch astronomer couple Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plate s taken by astronomer Tom Gehrels at the Palomar Observatory in California.", "requested_types": ["ASTRONOMICAL_OBJECT", "MISC", "LOCATION", "SCIENTIST", "ENZYME", "AWARD", "CHEMICAL_COMPOUND"], "distractors": ["AWARD", "ENZYME", "CHEMICAL_COMPOUND"], "gold": {"ASTRONOMICAL_OBJECT": ["Lycomedes"], "MISC": ["Dutch"], "LOCATION": ["Leiden", "Palomar Observatory", "California"], "SCIENTIST": ["Ingrid van Houten-Groeneveld", "Cornelis van Houten", "Tom Gehrels"], "ENZYME": [], "AWARD": [], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, LOCATION, SCIENTIST, ENZYME, AWARD, CHEMICAL_COMPOUND\n\nText:\nLycomedes was discovered on 26 September 1960, by Dutch astronomer couple Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plate s taken by astronomer Tom Gehrels at the Palomar Observatory in California.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, LOCATION, SCIENTIST, ENZYME, AWARD, CHEMICAL_COMPOUND\n\nText:\nLycomedes was discovered on 26 September 1960, by Dutch astronomer couple Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plate s taken by astronomer Tom Gehrels at the Palomar Observatory in California.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_87", "text": "In recognition of his contribution to the creation of modern electrical science, an international convention, signed at the 1881 International Exposition of Electricity, established the ampere as a standard unit of electrical measurement, along with the coulomb, volt, ohm, and watt, which are named, respectively, after Ampère's contemporaries Charles-Augustin de Coulomb of France, Alessandro Volta of Italy, Georg Ohm of Germany, and James Watt of Scotland.", "requested_types": ["COUNTRY", "MISC", "DISCIPLINE", "ORGANISATION", "EVENT", "SCIENTIST", "ACADEMIC_JOURNAL"], "distractors": ["ORGANISATION", "ACADEMIC_JOURNAL"], "gold": {"COUNTRY": ["France", "Italy", "Germany", "Scotland"], "MISC": ["ampere", "coulomb", "volt", "ohm", "watt"], "DISCIPLINE": ["electrical science"], "ORGANISATION": [], "EVENT": ["1881 International Exposition of Electricity"], "SCIENTIST": ["Ampère", "Charles-Augustin de Coulomb", "Alessandro Volta", "Georg Ohm", "James Watt"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, MISC, DISCIPLINE, ORGANISATION, EVENT, SCIENTIST, ACADEMIC_JOURNAL\n\nText:\nIn recognition of his contribution to the creation of modern electrical science, an international convention, signed at the 1881 International Exposition of Electricity, established the ampere as a standard unit of electrical measurement, along with the coulomb, volt, ohm, and watt, which are named, respectively, after Ampère's contemporaries Charles-Augustin de Coulomb of France, Alessandro Volta of Italy, Georg Ohm of Germany, and James Watt of Scotland.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, MISC, DISCIPLINE, ORGANISATION, EVENT, SCIENTIST, ACADEMIC_JOURNAL\n\nText:\nIn recognition of his contribution to the creation of modern electrical science, an international convention, signed at the 1881 International Exposition of Electricity, established the ampere as a standard unit of electrical measurement, along with the coulomb, volt, ohm, and watt, which are named, respectively, after Ampère's contemporaries Charles-Augustin de Coulomb of France, Alessandro Volta of Italy, Georg Ohm of Germany, and James Watt of Scotland.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_88", "text": "Five out of Finland's 15 universities ref group = note These are Aalto University, the University of Helsinki, the University of the Arts Helsinki, the Hanken School of Economics, and the National Defence University.", "requested_types": ["UNIVERSITY", "DISCIPLINE", "MISC", "LOCATION", "PERSON"], "distractors": ["LOCATION", "MISC", "DISCIPLINE"], "gold": {"UNIVERSITY": ["Aalto University", "University of Helsinki", "University of the Arts Helsinki", "Hanken School of Economics", "National Defence University"], "DISCIPLINE": [], "MISC": [], "LOCATION": [], "PERSON": ["Finland"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, DISCIPLINE, MISC, LOCATION, PERSON\n\nText:\nFive out of Finland's 15 universities ref group = note These are Aalto University, the University of Helsinki, the University of the Arts Helsinki, the Hanken School of Economics, and the National Defence University.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, DISCIPLINE, MISC, LOCATION, PERSON\n\nText:\nFive out of Finland's 15 universities ref group = note These are Aalto University, the University of Helsinki, the University of the Arts Helsinki, the Hanken School of Economics, and the National Defence University.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_89", "text": "The extended double stranded adapters are cleaved by Restriction enzyme at a specific restriction site located at 3 side of the tag sequence and will results in a 3 -dT overhang that will be ligated to the 3 -dA overhang on DNA libraries in adapter ligation step (Figure 1).", "requested_types": ["ORGANISATION", "MISC", "COUNTRY", "ENZYME"], "distractors": ["COUNTRY", "ORGANISATION"], "gold": {"ORGANISATION": [], "MISC": ["3 -dT overhang", "3 -dA overhang", "DNA libraries", "adapter ligation"], "COUNTRY": [], "ENZYME": ["Restriction enzyme"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, COUNTRY, ENZYME\n\nText:\nThe extended double stranded adapters are cleaved by Restriction enzyme at a specific restriction site located at 3 side of the tag sequence and will results in a 3 -dT overhang that will be ligated to the 3 -dA overhang on DNA libraries in adapter ligation step (Figure 1).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, COUNTRY, ENZYME\n\nText:\nThe extended double stranded adapters are cleaved by Restriction enzyme at a specific restriction site located at 3 side of the tag sequence and will results in a 3 -dT overhang that will be ligated to the 3 -dA overhang on DNA libraries in adapter ligation step (Figure 1).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_90", "text": "50000 Quaoar and 90377 Sedna are two Solar System objects discovered in this way by Michael E. Brown and others at Caltech using the Palomar Observatory 's Samuel Oschin telescope of and the Palomar-Quest large-area CCD camera.", "requested_types": ["ORGANISATION", "UNIVERSITY", "LOCATION", "MISC", "AWARD", "ASTRONOMICAL_OBJECT", "SCIENTIST"], "distractors": ["AWARD", "ORGANISATION"], "gold": {"ORGANISATION": [], "UNIVERSITY": ["Caltech"], "LOCATION": ["Palomar Observatory"], "MISC": ["Solar System", "Samuel Oschin telescope", "Palomar-Quest large-area CCD camera"], "AWARD": [], "ASTRONOMICAL_OBJECT": ["50000 Quaoar", "90377 Sedna"], "SCIENTIST": ["Michael E. Brown"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, UNIVERSITY, LOCATION, MISC, AWARD, ASTRONOMICAL_OBJECT, SCIENTIST\n\nText:\n50000 Quaoar and 90377 Sedna are two Solar System objects discovered in this way by Michael E. Brown and others at Caltech using the Palomar Observatory 's Samuel Oschin telescope of and the Palomar-Quest large-area CCD camera.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, UNIVERSITY, LOCATION, MISC, AWARD, ASTRONOMICAL_OBJECT, SCIENTIST\n\nText:\n50000 Quaoar and 90377 Sedna are two Solar System objects discovered in this way by Michael E. Brown and others at Caltech using the Palomar Observatory 's Samuel Oschin telescope of and the Palomar-Quest large-area CCD camera.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_91", "text": "He represented his country at the 2017 World Championships in Athletics without reaching the semifinals.", "requested_types": ["COUNTRY", "EVENT"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "EVENT": ["2017 World Championships in Athletics"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, EVENT\n\nText:\nHe represented his country at the 2017 World Championships in Athletics without reaching the semifinals.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, EVENT\n\nText:\nHe represented his country at the 2017 World Championships in Athletics without reaching the semifinals.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_92", "text": "The German university system allowed students to move easily from one university to another, so he spent summer semesters at Heidelberg University in 1902 and the University of Zurich in 1903.", "requested_types": ["ORGANISATION", "UNIVERSITY", "CHEMICAL_COMPOUND", "DISCIPLINE"], "distractors": ["ORGANISATION", "DISCIPLINE", "CHEMICAL_COMPOUND"], "gold": {"ORGANISATION": [], "UNIVERSITY": ["The German university", "Heidelberg University", "University of Zurich"], "CHEMICAL_COMPOUND": [], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, UNIVERSITY, CHEMICAL_COMPOUND, DISCIPLINE\n\nText:\nThe German university system allowed students to move easily from one university to another, so he spent summer semesters at Heidelberg University in 1902 and the University of Zurich in 1903.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, UNIVERSITY, CHEMICAL_COMPOUND, DISCIPLINE\n\nText:\nThe German university system allowed students to move easily from one university to another, so he spent summer semesters at Heidelberg University in 1902 and the University of Zurich in 1903.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_93", "text": "The apparently regulated genes encode subunits of RNA polymerase.", "requested_types": ["ORGANISATION", "EVENT", "ENZYME"], "distractors": ["ORGANISATION", "EVENT"], "gold": {"ORGANISATION": [], "EVENT": [], "ENZYME": ["RNA polymerase"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, EVENT, ENZYME\n\nText:\nThe apparently regulated genes encode subunits of RNA polymerase.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, EVENT, ENZYME\n\nText:\nThe apparently regulated genes encode subunits of RNA polymerase.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_94", "text": "STAT6, IRF4, GATA3 are absolutely required for TsubH / sub9 cell development and other such as PU.1, BATF, NF-κB, NFAT1, STAT5, AP-1 transcription factor contribute to TsubH / sub9 sub-population commitment and to IL-9 production.", "requested_types": ["MISC", "PROTEIN", "CHEMICAL_ELEMENT", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND", "CHEMICAL_ELEMENT"], "gold": {"MISC": ["STAT6", "GATA3", "TsubH / sub9 cell", "BATF", "NFAT1", "AP-1 transcription factor", "TsubH / sub9", "IL-9"], "PROTEIN": ["IRF4", "PU.1", "NF-κB", "STAT5"], "CHEMICAL_ELEMENT": [], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, CHEMICAL_ELEMENT, CHEMICAL_COMPOUND\n\nText:\nSTAT6, IRF4, GATA3 are absolutely required for TsubH / sub9 cell development and other such as PU.1, BATF, NF-κB, NFAT1, STAT5, AP-1 transcription factor contribute to TsubH / sub9 sub-population commitment and to IL-9 production.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, CHEMICAL_ELEMENT, CHEMICAL_COMPOUND\n\nText:\nSTAT6, IRF4, GATA3 are absolutely required for TsubH / sub9 cell development and other such as PU.1, BATF, NF-κB, NFAT1, STAT5, AP-1 transcription factor contribute to TsubH / sub9 sub-population commitment and to IL-9 production.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_95", "text": "Reverse transcriptase protocols at the time had difficulties with the secondary structure of mRNA, leading to abbreviated cDNAs that were difficult to align and invited further complications in downstream analysis.", "requested_types": ["ACADEMIC_JOURNAL", "EVENT", "ENZYME", "CHEMICAL_ELEMENT", "MISC"], "distractors": ["EVENT", "CHEMICAL_ELEMENT", "ACADEMIC_JOURNAL"], "gold": {"ACADEMIC_JOURNAL": [], "EVENT": [], "ENZYME": ["Reverse transcriptase"], "CHEMICAL_ELEMENT": [], "MISC": ["mRNA", "cDNAs"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, EVENT, ENZYME, CHEMICAL_ELEMENT, MISC\n\nText:\nReverse transcriptase protocols at the time had difficulties with the secondary structure of mRNA, leading to abbreviated cDNAs that were difficult to align and invited further complications in downstream analysis.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, EVENT, ENZYME, CHEMICAL_ELEMENT, MISC\n\nText:\nReverse transcriptase protocols at the time had difficulties with the secondary structure of mRNA, leading to abbreviated cDNAs that were difficult to align and invited further complications in downstream analysis.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_96", "text": "These include the Naval Base San Diego, Marine Corps Base Camp Pendleton, Marine Corps Air Station Miramar, and Naval Air Station North Island.", "requested_types": ["LOCATION", "ORGANISATION", "PERSON"], "distractors": ["PERSON", "ORGANISATION"], "gold": {"LOCATION": ["Naval Base San Diego", "Marine Corps Base Camp Pendleton", "Marine Corps Air Station Miramar", "Naval Air Station North Island"], "ORGANISATION": [], "PERSON": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ORGANISATION, PERSON\n\nText:\nThese include the Naval Base San Diego, Marine Corps Base Camp Pendleton, Marine Corps Air Station Miramar, and Naval Air Station North Island.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ORGANISATION, PERSON\n\nText:\nThese include the Naval Base San Diego, Marine Corps Base Camp Pendleton, Marine Corps Air Station Miramar, and Naval Air Station North Island.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_97", "text": "Drewry Ice Stream in Ellsworth Land and Mount Drewry in the Queen Alexandra Range are named after him.", "requested_types": ["ACADEMIC_JOURNAL", "LOCATION", "AWARD"], "distractors": ["AWARD", "ACADEMIC_JOURNAL"], "gold": {"ACADEMIC_JOURNAL": [], "LOCATION": ["Drewry Ice Stream", "Ellsworth Land", "Mount Drewry", "Queen Alexandra Range"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, LOCATION, AWARD\n\nText:\nDrewry Ice Stream in Ellsworth Land and Mount Drewry in the Queen Alexandra Range are named after him.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, LOCATION, AWARD\n\nText:\nDrewry Ice Stream in Ellsworth Land and Mount Drewry in the Queen Alexandra Range are named after him.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_98", "text": "He was a life member of the Franklin Institute, the National Academy of Engineering and the Society for Advancement of Management.", "requested_types": ["ORGANISATION", "THEORY", "COUNTRY"], "distractors": ["THEORY", "COUNTRY"], "gold": {"ORGANISATION": ["Franklin Institute", "National Academy of Engineering", "Society for Advancement of Management"], "THEORY": [], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, THEORY, COUNTRY\n\nText:\nHe was a life member of the Franklin Institute, the National Academy of Engineering and the Society for Advancement of Management.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, THEORY, COUNTRY\n\nText:\nHe was a life member of the Franklin Institute, the National Academy of Engineering and the Society for Advancement of Management.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_99", "text": "The principal candidates for the origin of Mars' methane include non-biological processes such as Properties of water -rock reactions, radiolysis of water, and pyrite formation, all of which produce Hydrogen that could then generate methane and other hydrocarbons via Fischer-Tropsch synthesis with Carbon monoxide and COsub2 / sub.", "requested_types": ["CHEMICAL_COMPOUND", "MISC", "ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT", "DISCIPLINE"], "distractors": ["DISCIPLINE"], "gold": {"CHEMICAL_COMPOUND": ["methane", "hydrocarbons", "Carbon monoxide", "COsub2 / sub"], "MISC": ["water -rock reactions", "radiolysis of water", "pyrite formation", "Fischer-Tropsch synthesis"], "ASTRONOMICAL_OBJECT": ["Mars"], "CHEMICAL_ELEMENT": ["Hydrogen"], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, MISC, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, DISCIPLINE\n\nText:\nThe principal candidates for the origin of Mars' methane include non-biological processes such as Properties of water -rock reactions, radiolysis of water, and pyrite formation, all of which produce Hydrogen that could then generate methane and other hydrocarbons via Fischer-Tropsch synthesis with Carbon monoxide and COsub2 / sub.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, MISC, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, DISCIPLINE\n\nText:\nThe principal candidates for the origin of Mars' methane include non-biological processes such as Properties of water -rock reactions, radiolysis of water, and pyrite formation, all of which produce Hydrogen that could then generate methane and other hydrocarbons via Fischer-Tropsch synthesis with Carbon monoxide and COsub2 / sub.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_100", "text": "Such a definition of the term planet could also have led to changes in classification for the trans-Neptunian object s, and the asteroids 4 Vesta, 2 Pallas, and 10 Hygiea.", "requested_types": ["MISC", "ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND"], "gold": {"MISC": ["trans-Neptunian object"], "ASTRONOMICAL_OBJECT": ["4 Vesta", "2 Pallas", "10 Hygiea"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND\n\nText:\nSuch a definition of the term planet could also have led to changes in classification for the trans-Neptunian object s, and the asteroids 4 Vesta, 2 Pallas, and 10 Hygiea.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND\n\nText:\nSuch a definition of the term planet could also have led to changes in classification for the trans-Neptunian object s, and the asteroids 4 Vesta, 2 Pallas, and 10 Hygiea.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_101", "text": "The minor planets 1062 Ljuba and 1086 Nata were also named after Soviet female paratroopers Lyuba Berlin (1915-1936) and Nata Babushkina (1915-1936), respectively.", "requested_types": ["UNIVERSITY", "MISC", "ASTRONOMICAL_OBJECT", "PERSON", "SCIENTIST", "COUNTRY"], "distractors": ["UNIVERSITY", "COUNTRY", "SCIENTIST"], "gold": {"UNIVERSITY": [], "MISC": ["minor planets", "Soviet"], "ASTRONOMICAL_OBJECT": ["1062 Ljuba", "1086 Nata"], "PERSON": ["Lyuba Berlin", "Nata Babushkina"], "SCIENTIST": [], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, MISC, ASTRONOMICAL_OBJECT, PERSON, SCIENTIST, COUNTRY\n\nText:\nThe minor planets 1062 Ljuba and 1086 Nata were also named after Soviet female paratroopers Lyuba Berlin (1915-1936) and Nata Babushkina (1915-1936), respectively.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, MISC, ASTRONOMICAL_OBJECT, PERSON, SCIENTIST, COUNTRY\n\nText:\nThe minor planets 1062 Ljuba and 1086 Nata were also named after Soviet female paratroopers Lyuba Berlin (1915-1936) and Nata Babushkina (1915-1936), respectively.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_102", "text": "This minor planet was named for Aesculapius, the Greek and Roman demigod of medicine and healing, son of Apollo and Coronis, after whom the asteroids 158 Koronis and 1862 Apollo are named, respectively.", "requested_types": ["PERSON", "ACADEMIC_JOURNAL", "LOCATION", "ASTRONOMICAL_OBJECT", "PROTEIN", "MISC", "DISCIPLINE"], "distractors": ["ACADEMIC_JOURNAL", "PROTEIN", "LOCATION"], "gold": {"PERSON": ["Apollo", "Coronis"], "ACADEMIC_JOURNAL": [], "LOCATION": [], "ASTRONOMICAL_OBJECT": ["Aesculapius", "158 Koronis", "1862 Apollo"], "PROTEIN": [], "MISC": ["minor planet", "Greek", "Roman"], "DISCIPLINE": ["medicine"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ACADEMIC_JOURNAL, LOCATION, ASTRONOMICAL_OBJECT, PROTEIN, MISC, DISCIPLINE\n\nText:\nThis minor planet was named for Aesculapius, the Greek and Roman demigod of medicine and healing, son of Apollo and Coronis, after whom the asteroids 158 Koronis and 1862 Apollo are named, respectively.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ACADEMIC_JOURNAL, LOCATION, ASTRONOMICAL_OBJECT, PROTEIN, MISC, DISCIPLINE\n\nText:\nThis minor planet was named for Aesculapius, the Greek and Roman demigod of medicine and healing, son of Apollo and Coronis, after whom the asteroids 158 Koronis and 1862 Apollo are named, respectively.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_103", "text": "In this publication he proposed the names still used today for the seven then-known satellites of Saturn: Mimas, Enceladus, Tethys, Dione, Rhea, Titan, and Iapetus.", "requested_types": ["ASTRONOMICAL_OBJECT", "DISCIPLINE"], "distractors": ["DISCIPLINE"], "gold": {"ASTRONOMICAL_OBJECT": ["Saturn", "Mimas", "Enceladus", "Tethys", "Dione", "Rhea", "Titan", "Iapetus"], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, DISCIPLINE\n\nText:\nIn this publication he proposed the names still used today for the seven then-known satellites of Saturn: Mimas, Enceladus, Tethys, Dione, Rhea, Titan, and Iapetus.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, DISCIPLINE\n\nText:\nIn this publication he proposed the names still used today for the seven then-known satellites of Saturn: Mimas, Enceladus, Tethys, Dione, Rhea, Titan, and Iapetus.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_104", "text": "It is probably the fifth- most-massive asteroid after Ceres, 4 Vesta, 2 Pallas, and 10 Hygiea, with a mass estimated to be 1.2% of the mass of the entire asteroid belt.", "requested_types": ["ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT", "MISC", "CHEMICAL_COMPOUND", "PROTEIN"], "distractors": ["CHEMICAL_COMPOUND", "PROTEIN", "CHEMICAL_ELEMENT"], "gold": {"ASTRONOMICAL_OBJECT": ["Ceres", "4 Vesta", "2 Pallas", "10 Hygiea"], "CHEMICAL_ELEMENT": [], "MISC": ["asteroid belt"], "CHEMICAL_COMPOUND": [], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, MISC, CHEMICAL_COMPOUND, PROTEIN\n\nText:\nIt is probably the fifth- most-massive asteroid after Ceres, 4 Vesta, 2 Pallas, and 10 Hygiea, with a mass estimated to be 1.2% of the mass of the entire asteroid belt.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, MISC, CHEMICAL_COMPOUND, PROTEIN\n\nText:\nIt is probably the fifth- most-massive asteroid after Ceres, 4 Vesta, 2 Pallas, and 10 Hygiea, with a mass estimated to be 1.2% of the mass of the entire asteroid belt.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_105", "text": "They were proposed independently by Alfred J. Lotka in 1925 and Vito Volterra in 1926.", "requested_types": ["ENZYME", "SCIENTIST", "PROTEIN", "UNIVERSITY"], "distractors": ["ENZYME", "PROTEIN", "UNIVERSITY"], "gold": {"ENZYME": [], "SCIENTIST": ["Alfred J. Lotka", "Vito Volterra"], "PROTEIN": [], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, SCIENTIST, PROTEIN, UNIVERSITY\n\nText:\nThey were proposed independently by Alfred J. Lotka in 1925 and Vito Volterra in 1926.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, SCIENTIST, PROTEIN, UNIVERSITY\n\nText:\nThey were proposed independently by Alfred J. Lotka in 1925 and Vito Volterra in 1926.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_106", "text": "Lolitrem B affects a kind of ion channel called BK channel s.", "requested_types": ["PERSON", "CHEMICAL_ELEMENT", "CHEMICAL_COMPOUND", "DISCIPLINE", "PROTEIN", "MISC"], "distractors": ["DISCIPLINE", "CHEMICAL_ELEMENT", "PERSON"], "gold": {"PERSON": [], "CHEMICAL_ELEMENT": [], "CHEMICAL_COMPOUND": ["Lolitrem B"], "DISCIPLINE": [], "PROTEIN": ["BK channel"], "MISC": ["ion channel"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, CHEMICAL_ELEMENT, CHEMICAL_COMPOUND, DISCIPLINE, PROTEIN, MISC\n\nText:\nLolitrem B affects a kind of ion channel called BK channel s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, CHEMICAL_ELEMENT, CHEMICAL_COMPOUND, DISCIPLINE, PROTEIN, MISC\n\nText:\nLolitrem B affects a kind of ion channel called BK channel s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_107", "text": "Closer to the city, several sporting complexes are located, including the Hobart Aquatic Centre, Hobart International Tennis Centre the Domain Athletic Centre, and the Hobart TCA Ground.", "requested_types": ["ORGANISATION", "LOCATION"], "distractors": ["ORGANISATION"], "gold": {"ORGANISATION": [], "LOCATION": ["Hobart Aquatic Centre", "Hobart International Tennis Centre", "Domain Athletic Centre", "Hobart TCA Ground"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, LOCATION\n\nText:\nCloser to the city, several sporting complexes are located, including the Hobart Aquatic Centre, Hobart International Tennis Centre the Domain Athletic Centre, and the Hobart TCA Ground.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, LOCATION\n\nText:\nCloser to the city, several sporting complexes are located, including the Hobart Aquatic Centre, Hobart International Tennis Centre the Domain Athletic Centre, and the Hobart TCA Ground.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_108", "text": "Additionally, she has received five Academy of Country Music Awards, six Country Music Association Awards, and a Grammy Award.", "requested_types": ["ASTRONOMICAL_OBJECT", "EVENT", "CHEMICAL_COMPOUND", "AWARD"], "distractors": ["ASTRONOMICAL_OBJECT", "EVENT", "CHEMICAL_COMPOUND"], "gold": {"ASTRONOMICAL_OBJECT": [], "EVENT": [], "CHEMICAL_COMPOUND": [], "AWARD": ["Academy of Country Music Awards", "Country Music Association Awards", "Grammy Award"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, EVENT, CHEMICAL_COMPOUND, AWARD\n\nText:\nAdditionally, she has received five Academy of Country Music Awards, six Country Music Association Awards, and a Grammy Award.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, EVENT, CHEMICAL_COMPOUND, AWARD\n\nText:\nAdditionally, she has received five Academy of Country Music Awards, six Country Music Association Awards, and a Grammy Award.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_109", "text": "Juilfs was a theoretical physics assistant from 1938 to 1945 at the Kaiser-Wilhelm Institut für Physik (KWIP, Kaiser Wilhelm Institute for Physics; today, the Max-Planck Institut für Physik), first for Max von Laue and from 1943 to Werner Heisenberg and Hentschel, 1996, Appendix F.", "requested_types": ["AWARD", "ACADEMIC_JOURNAL", "ORGANISATION", "DISCIPLINE", "SCIENTIST"], "distractors": ["AWARD", "DISCIPLINE", "ACADEMIC_JOURNAL"], "gold": {"AWARD": [], "ACADEMIC_JOURNAL": [], "ORGANISATION": ["Kaiser-Wilhelm Institut für Physik", "KWIP", "Kaiser Wilhelm Institute for Physics", "Max-Planck Institut für Physik"], "DISCIPLINE": [], "SCIENTIST": ["Juilfs", "Max von Laue", "Werner Heisenberg", "Hentschel"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ACADEMIC_JOURNAL, ORGANISATION, DISCIPLINE, SCIENTIST\n\nText:\nJuilfs was a theoretical physics assistant from 1938 to 1945 at the Kaiser-Wilhelm Institut für Physik (KWIP, Kaiser Wilhelm Institute for Physics; today, the Max-Planck Institut für Physik), first for Max von Laue and from 1943 to Werner Heisenberg and Hentschel, 1996, Appendix F.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ACADEMIC_JOURNAL, ORGANISATION, DISCIPLINE, SCIENTIST\n\nText:\nJuilfs was a theoretical physics assistant from 1938 to 1945 at the Kaiser-Wilhelm Institut für Physik (KWIP, Kaiser Wilhelm Institute for Physics; today, the Max-Planck Institut für Physik), first for Max von Laue and from 1943 to Werner Heisenberg and Hentschel, 1996, Appendix F.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_110", "text": "Excessive amounts of chlorine in the buffer should also be avoided, since this will overlap with the sulfur peak; Potassium bromide and Sodium bromide are suitable alternatives.", "requested_types": ["MISC", "CHEMICAL_COMPOUND", "CHEMICAL_ELEMENT", "ACADEMIC_JOURNAL"], "distractors": ["ACADEMIC_JOURNAL", "MISC"], "gold": {"MISC": [], "CHEMICAL_COMPOUND": ["Potassium bromide", "Sodium bromide"], "CHEMICAL_ELEMENT": ["chlorine", "sulfur"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_COMPOUND, CHEMICAL_ELEMENT, ACADEMIC_JOURNAL\n\nText:\nExcessive amounts of chlorine in the buffer should also be avoided, since this will overlap with the sulfur peak; Potassium bromide and Sodium bromide are suitable alternatives.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_COMPOUND, CHEMICAL_ELEMENT, ACADEMIC_JOURNAL\n\nText:\nExcessive amounts of chlorine in the buffer should also be avoided, since this will overlap with the sulfur peak; Potassium bromide and Sodium bromide are suitable alternatives.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_111", "text": "The original body farm is the University of Tennessee Anthropological Research Facility located a few miles from downtown on Alcoa Highway in Tennessee, behind the University of Tennessee Medical Center.", "requested_types": ["UNIVERSITY", "LOCATION", "SCIENTIST"], "distractors": ["SCIENTIST"], "gold": {"UNIVERSITY": ["University of Tennessee Anthropological Research Facility", "University of Tennessee Medical Center"], "LOCATION": ["Alcoa Highway in", "Tennessee"], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, LOCATION, SCIENTIST\n\nText:\nThe original body farm is the University of Tennessee Anthropological Research Facility located a few miles from downtown on Alcoa Highway in Tennessee, behind the University of Tennessee Medical Center.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, LOCATION, SCIENTIST\n\nText:\nThe original body farm is the University of Tennessee Anthropological Research Facility located a few miles from downtown on Alcoa Highway in Tennessee, behind the University of Tennessee Medical Center.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_112", "text": "These experimental and mathematical analyses were applied to several areas of geophysics: Earth's shape, density, and gravity field (Pierre Bouguer, Alexis Clairaut and Henry Cavendish), Earth's magnetic field (Alexander von Humboldt, Edmund Halley and Carl Friedrich Gauss), seismology (John Milne and Robert Mallet), and the Earth's age, heat and radioactivity (Arthur Holmes and William Thomson, 1st Baron Kelvin).", "requested_types": ["LOCATION", "DISCIPLINE", "MISC", "CHEMICAL_COMPOUND", "ASTRONOMICAL_OBJECT", "SCIENTIST"], "distractors": ["LOCATION", "CHEMICAL_COMPOUND"], "gold": {"LOCATION": [], "DISCIPLINE": ["geophysics", "seismology"], "MISC": ["mathematical"], "CHEMICAL_COMPOUND": [], "ASTRONOMICAL_OBJECT": ["Earth"], "SCIENTIST": ["Pierre Bouguer", "Alexis Clairaut", "Henry Cavendish", "Alexander von Humboldt", "Edmund Halley", "Carl Friedrich Gauss", "John Milne", "Robert Mallet", "Arthur Holmes", "William Thomson, 1st Baron Kelvin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, DISCIPLINE, MISC, CHEMICAL_COMPOUND, ASTRONOMICAL_OBJECT, SCIENTIST\n\nText:\nThese experimental and mathematical analyses were applied to several areas of geophysics: Earth's shape, density, and gravity field (Pierre Bouguer, Alexis Clairaut and Henry Cavendish), Earth's magnetic field (Alexander von Humboldt, Edmund Halley and Carl Friedrich Gauss), seismology (John Milne and Robert Mallet), and the Earth's age, heat and radioactivity (Arthur Holmes and William Thomson, 1st Baron Kelvin).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, DISCIPLINE, MISC, CHEMICAL_COMPOUND, ASTRONOMICAL_OBJECT, SCIENTIST\n\nText:\nThese experimental and mathematical analyses were applied to several areas of geophysics: Earth's shape, density, and gravity field (Pierre Bouguer, Alexis Clairaut and Henry Cavendish), Earth's magnetic field (Alexander von Humboldt, Edmund Halley and Carl Friedrich Gauss), seismology (John Milne and Robert Mallet), and the Earth's age, heat and radioactivity (Arthur Holmes and William Thomson, 1st Baron Kelvin).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_113", "text": "The three required enzyme activities are: exonuclease, DNA polymerase, and DNA ligase.", "requested_types": ["ENZYME", "LOCATION"], "distractors": ["LOCATION"], "gold": {"ENZYME": ["exonuclease", "DNA polymerase", "DNA ligase"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, LOCATION\n\nText:\nThe three required enzyme activities are: exonuclease, DNA polymerase, and DNA ligase.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, LOCATION\n\nText:\nThe three required enzyme activities are: exonuclease, DNA polymerase, and DNA ligase.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_114", "text": "Computer programs that solve these problems have been used to research a broad range of scientific topics from Adenosine diphosphate to breast cancer.", "requested_types": ["CHEMICAL_COMPOUND", "MISC", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_COMPOUND": ["Adenosine diphosphate"], "MISC": ["breast cancer"], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, MISC, CHEMICAL_ELEMENT\n\nText:\nComputer programs that solve these problems have been used to research a broad range of scientific topics from Adenosine diphosphate to breast cancer.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, MISC, CHEMICAL_ELEMENT\n\nText:\nComputer programs that solve these problems have been used to research a broad range of scientific topics from Adenosine diphosphate to breast cancer.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_115", "text": "Observations are currently being coordinated by the Association of Lunar and Planetary Observers and the British Astronomical Association to re-observe sites where transient lunar phenomena were reported in the past.", "requested_types": ["LOCATION", "ENZYME", "ORGANISATION", "PERSON"], "distractors": ["LOCATION", "PERSON", "ENZYME"], "gold": {"LOCATION": [], "ENZYME": [], "ORGANISATION": ["Association of Lunar and Planetary Observers", "British Astronomical Association"], "PERSON": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ENZYME, ORGANISATION, PERSON\n\nText:\nObservations are currently being coordinated by the Association of Lunar and Planetary Observers and the British Astronomical Association to re-observe sites where transient lunar phenomena were reported in the past.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ENZYME, ORGANISATION, PERSON\n\nText:\nObservations are currently being coordinated by the Association of Lunar and Planetary Observers and the British Astronomical Association to re-observe sites where transient lunar phenomena were reported in the past.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_116", "text": "The programs described are Arthur Samuel 's checkers program for the IBM 701, Daniel Bobrow' s STUDENT, Allen Newell and Herbert A. Simon 's Logic Theorist and Terry Winograd' s SHRDLU.", "requested_types": ["SCIENTIST", "EVENT", "AWARD", "UNIVERSITY", "MISC"], "distractors": ["UNIVERSITY", "EVENT", "AWARD"], "gold": {"SCIENTIST": ["Arthur Samuel", "Daniel Bobrow", "Allen Newell", "Herbert A. Simon", "Terry Winograd"], "EVENT": [], "AWARD": [], "UNIVERSITY": [], "MISC": ["IBM 701", "STUDENT", "Logic Theorist", "SHRDLU"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, EVENT, AWARD, UNIVERSITY, MISC\n\nText:\nThe programs described are Arthur Samuel 's checkers program for the IBM 701, Daniel Bobrow' s STUDENT, Allen Newell and Herbert A. Simon 's Logic Theorist and Terry Winograd' s SHRDLU.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, EVENT, AWARD, UNIVERSITY, MISC\n\nText:\nThe programs described are Arthur Samuel 's checkers program for the IBM 701, Daniel Bobrow' s STUDENT, Allen Newell and Herbert A. Simon 's Logic Theorist and Terry Winograd' s SHRDLU.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_117", "text": "All of these low-numbered asteroids have numbers between and were discovered between 1876 and the 1930s, predominantly by astronomers Auguste Charlois, Johann Palisa, Max Wolf and Karl Reinmuth (also see category).", "requested_types": ["EVENT", "SCIENTIST", "AWARD", "ENZYME"], "distractors": ["EVENT", "AWARD", "ENZYME"], "gold": {"EVENT": [], "SCIENTIST": ["Auguste Charlois", "Johann Palisa", "Max Wolf", "Karl Reinmuth"], "AWARD": [], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, SCIENTIST, AWARD, ENZYME\n\nText:\nAll of these low-numbered asteroids have numbers between and were discovered between 1876 and the 1930s, predominantly by astronomers Auguste Charlois, Johann Palisa, Max Wolf and Karl Reinmuth (also see category).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, SCIENTIST, AWARD, ENZYME\n\nText:\nAll of these low-numbered asteroids have numbers between and were discovered between 1876 and the 1930s, predominantly by astronomers Auguste Charlois, Johann Palisa, Max Wolf and Karl Reinmuth (also see category).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_118", "text": "For herbivore mammal species, DNA is usually extracted from faeces samples Potentially, this application is applicable to all herbivorous species feeding on angiosperms and gymnosperms Alternatively to the trn L aprroach, the markers RuBisCO, ITS2, Maturase K, trnH-psbA can be used to amplify plant species.", "requested_types": ["EVENT", "MISC", "CHEMICAL_ELEMENT", "ENZYME", "ORGANISATION"], "distractors": ["EVENT", "CHEMICAL_ELEMENT", "ORGANISATION"], "gold": {"EVENT": [], "MISC": ["DNA", "angiosperms", "gymnosperms", "ITS2", "Maturase K", "trnH-psbA"], "CHEMICAL_ELEMENT": [], "ENZYME": ["RuBisCO"], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, MISC, CHEMICAL_ELEMENT, ENZYME, ORGANISATION\n\nText:\nFor herbivore mammal species, DNA is usually extracted from faeces samples Potentially, this application is applicable to all herbivorous species feeding on angiosperms and gymnosperms Alternatively to the trn L aprroach, the markers RuBisCO, ITS2, Maturase K, trnH-psbA can be used to amplify plant species.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, MISC, CHEMICAL_ELEMENT, ENZYME, ORGANISATION\n\nText:\nFor herbivore mammal species, DNA is usually extracted from faeces samples Potentially, this application is applicable to all herbivorous species feeding on angiosperms and gymnosperms Alternatively to the trn L aprroach, the markers RuBisCO, ITS2, Maturase K, trnH-psbA can be used to amplify plant species.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_119", "text": "This is the Hong Kong roster in the women's basketball tournament of the 2018 Asian Games.", "requested_types": ["EVENT", "AWARD", "LOCATION", "MISC"], "distractors": ["AWARD"], "gold": {"EVENT": ["2018 Asian Games"], "AWARD": [], "LOCATION": ["Hong Kong"], "MISC": ["women's basketball tournament"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, AWARD, LOCATION, MISC\n\nText:\nThis is the Hong Kong roster in the women's basketball tournament of the 2018 Asian Games.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, AWARD, LOCATION, MISC\n\nText:\nThis is the Hong Kong roster in the women's basketball tournament of the 2018 Asian Games.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_120", "text": "It is probably the fifth- most-massive asteroid after Ceres, 4 Vesta, 2 Pallas, and 10 Hygiea, with a mass estimated to be 1.2% of the mass of the entire asteroid belt.", "requested_types": ["ASTRONOMICAL_OBJECT", "UNIVERSITY"], "distractors": ["UNIVERSITY"], "gold": {"ASTRONOMICAL_OBJECT": ["Ceres", "4 Vesta", "2 Pallas", "10 Hygiea"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, UNIVERSITY\n\nText:\nIt is probably the fifth- most-massive asteroid after Ceres, 4 Vesta, 2 Pallas, and 10 Hygiea, with a mass estimated to be 1.2% of the mass of the entire asteroid belt.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, UNIVERSITY\n\nText:\nIt is probably the fifth- most-massive asteroid after Ceres, 4 Vesta, 2 Pallas, and 10 Hygiea, with a mass estimated to be 1.2% of the mass of the entire asteroid belt.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_121", "text": "Lucifer was the name used by the Romans for the Venus and the goddess Venus.", "requested_types": ["UNIVERSITY", "CHEMICAL_COMPOUND", "MISC", "ASTRONOMICAL_OBJECT"], "distractors": ["CHEMICAL_COMPOUND", "UNIVERSITY"], "gold": {"UNIVERSITY": [], "CHEMICAL_COMPOUND": [], "MISC": ["Romans"], "ASTRONOMICAL_OBJECT": ["Lucifer", "Venus"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, CHEMICAL_COMPOUND, MISC, ASTRONOMICAL_OBJECT\n\nText:\nLucifer was the name used by the Romans for the Venus and the goddess Venus.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, CHEMICAL_COMPOUND, MISC, ASTRONOMICAL_OBJECT\n\nText:\nLucifer was the name used by the Romans for the Venus and the goddess Venus.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_122", "text": "The ten attendees were conference organizer J. Peter Pearman, Frank Drake, Philip Morrison, businessman and radio amateur Dana Atchley, chemist Melvin Calvin, astronomer Su-Shu Huang, neuroscientist John C. Lilly, inventor Barney Oliver, astronomer Carl Sagan and radio-astronomer Otto Struve.", "requested_types": ["COUNTRY", "CHEMICAL_COMPOUND", "PERSON", "SCIENTIST"], "distractors": ["CHEMICAL_COMPOUND", "COUNTRY"], "gold": {"COUNTRY": [], "CHEMICAL_COMPOUND": [], "PERSON": ["Dana Atchley"], "SCIENTIST": ["J. Peter Pearman", "Frank Drake", "Philip Morrison", "Melvin Calvin", "Su-Shu Huang", "John C. Lilly", "Barney Oliver", "Carl Sagan", "Otto Struve"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, CHEMICAL_COMPOUND, PERSON, SCIENTIST\n\nText:\nThe ten attendees were conference organizer J. Peter Pearman, Frank Drake, Philip Morrison, businessman and radio amateur Dana Atchley, chemist Melvin Calvin, astronomer Su-Shu Huang, neuroscientist John C. Lilly, inventor Barney Oliver, astronomer Carl Sagan and radio-astronomer Otto Struve.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, CHEMICAL_COMPOUND, PERSON, SCIENTIST\n\nText:\nThe ten attendees were conference organizer J. Peter Pearman, Frank Drake, Philip Morrison, businessman and radio amateur Dana Atchley, chemist Melvin Calvin, astronomer Su-Shu Huang, neuroscientist John C. Lilly, inventor Barney Oliver, astronomer Carl Sagan and radio-astronomer Otto Struve.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_123", "text": "The planets of the Solar System are divided into two groups: the four inner planets are the terrestrial planet s (Mercury, Venus, Earth and Mars), with relatively small sizes and rocky surfaces.", "requested_types": ["MISC", "LOCATION", "EVENT", "ASTRONOMICAL_OBJECT", "ORGANISATION"], "distractors": ["EVENT", "LOCATION", "ORGANISATION"], "gold": {"MISC": ["Solar System", "terrestrial planet"], "LOCATION": [], "EVENT": [], "ASTRONOMICAL_OBJECT": ["Mercury", "Venus", "Earth", "Mars"], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, LOCATION, EVENT, ASTRONOMICAL_OBJECT, ORGANISATION\n\nText:\nThe planets of the Solar System are divided into two groups: the four inner planets are the terrestrial planet s (Mercury, Venus, Earth and Mars), with relatively small sizes and rocky surfaces.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, LOCATION, EVENT, ASTRONOMICAL_OBJECT, ORGANISATION\n\nText:\nThe planets of the Solar System are divided into two groups: the four inner planets are the terrestrial planet s (Mercury, Venus, Earth and Mars), with relatively small sizes and rocky surfaces.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_124", "text": "Recognition of Kilby's outstanding achievements have been made by the Institute of Electrical and Electronics Engineers (IEEE), including the election to IEEE Fellow in 1966, the IEEE David Sarnoff Award in 1966, was created in 1995.", "requested_types": ["AWARD", "SCIENTIST", "ORGANISATION", "THEORY"], "distractors": ["THEORY"], "gold": {"AWARD": ["IEEE Fellow", "IEEE David Sarnoff Award"], "SCIENTIST": ["Kilby"], "ORGANISATION": ["Institute of Electrical and Electronics Engineers", "IEEE"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, SCIENTIST, ORGANISATION, THEORY\n\nText:\nRecognition of Kilby's outstanding achievements have been made by the Institute of Electrical and Electronics Engineers (IEEE), including the election to IEEE Fellow in 1966, the IEEE David Sarnoff Award in 1966, was created in 1995.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, SCIENTIST, ORGANISATION, THEORY\n\nText:\nRecognition of Kilby's outstanding achievements have been made by the Institute of Electrical and Electronics Engineers (IEEE), including the election to IEEE Fellow in 1966, the IEEE David Sarnoff Award in 1966, was created in 1995.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_125", "text": "With a diameter of approximately 10 kilometers, Amundsenia is one of the largest mid-sized Mars-crossing asteroid s such as 1139 Atami (9.35 km), 1474 Beira (14.9 km), 1011 Laodamia (7.39 km), 1727 Mette (est 9 km), 1131 Porzia (7.13 km), 1235 Schorria (est.", "requested_types": ["ENZYME", "ASTRONOMICAL_OBJECT", "SCIENTIST", "MISC"], "distractors": ["SCIENTIST", "ENZYME"], "gold": {"ENZYME": [], "ASTRONOMICAL_OBJECT": ["Amundsenia", "1139 Atami", "1474 Beira", "1011 Laodamia", "1727 Mette", "1131 Porzia", "1235 Schorria"], "SCIENTIST": [], "MISC": ["Mars-crossing asteroid"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ASTRONOMICAL_OBJECT, SCIENTIST, MISC\n\nText:\nWith a diameter of approximately 10 kilometers, Amundsenia is one of the largest mid-sized Mars-crossing asteroid s such as 1139 Atami (9.35 km), 1474 Beira (14.9 km), 1011 Laodamia (7.39 km), 1727 Mette (est 9 km), 1131 Porzia (7.13 km), 1235 Schorria (est.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ASTRONOMICAL_OBJECT, SCIENTIST, MISC\n\nText:\nWith a diameter of approximately 10 kilometers, Amundsenia is one of the largest mid-sized Mars-crossing asteroid s such as 1139 Atami (9.35 km), 1474 Beira (14.9 km), 1011 Laodamia (7.39 km), 1727 Mette (est 9 km), 1131 Porzia (7.13 km), 1235 Schorria (est.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_126", "text": "Beyond the Earth's atmosphere, thunderstorms have also been observed on the planets of Jupiter, Saturn, Neptune, and, probably, Venus.", "requested_types": ["PROTEIN", "EVENT", "ASTRONOMICAL_OBJECT"], "distractors": ["EVENT", "PROTEIN"], "gold": {"PROTEIN": [], "EVENT": [], "ASTRONOMICAL_OBJECT": ["Earth", "Jupiter", "Saturn", "Neptune", "Venus"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, EVENT, ASTRONOMICAL_OBJECT\n\nText:\nBeyond the Earth's atmosphere, thunderstorms have also been observed on the planets of Jupiter, Saturn, Neptune, and, probably, Venus.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, EVENT, ASTRONOMICAL_OBJECT\n\nText:\nBeyond the Earth's atmosphere, thunderstorms have also been observed on the planets of Jupiter, Saturn, Neptune, and, probably, Venus.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_127", "text": "Such measures include proofreading by some DNA polymerase during replication, DNA mismatch repair following replication, DNA proofreading and repair.", "requested_types": ["PROTEIN", "EVENT", "CHEMICAL_COMPOUND", "ENZYME", "MISC"], "distractors": ["CHEMICAL_COMPOUND", "EVENT"], "gold": {"PROTEIN": ["DNA mismatch repair"], "EVENT": [], "CHEMICAL_COMPOUND": [], "ENZYME": ["DNA polymerase"], "MISC": ["DNA"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, EVENT, CHEMICAL_COMPOUND, ENZYME, MISC\n\nText:\nSuch measures include proofreading by some DNA polymerase during replication, DNA mismatch repair following replication, DNA proofreading and repair.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, EVENT, CHEMICAL_COMPOUND, ENZYME, MISC\n\nText:\nSuch measures include proofreading by some DNA polymerase during replication, DNA mismatch repair following replication, DNA proofreading and repair.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_128", "text": "New DNA synthesis by DNA polymerase s is one of the ways radiation induced DNA damage can be repaired.", "requested_types": ["PERSON", "UNIVERSITY", "MISC", "ENZYME"], "distractors": ["UNIVERSITY", "PERSON"], "gold": {"PERSON": [], "UNIVERSITY": [], "MISC": ["DNA synthesis", "DNA"], "ENZYME": ["DNA polymerase"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, UNIVERSITY, MISC, ENZYME\n\nText:\nNew DNA synthesis by DNA polymerase s is one of the ways radiation induced DNA damage can be repaired.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, UNIVERSITY, MISC, ENZYME\n\nText:\nNew DNA synthesis by DNA polymerase s is one of the ways radiation induced DNA damage can be repaired.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_129", "text": "CCR3 is a receptor for multiple inflammatory / inducible CC chemokines, including CCL11, CCL26, CCL7, CCL13, CCL15, CCL24 and CCL5 that attract eosinophils, and CCL28 that attracts B and T lymphocyte s to mucosal tissue s.ref name = Youn / It is most highly expressed in both eosinophils and basophils, but can also be found in Th1 and Th2 cells and airway epithelial cells.", "requested_types": ["PROTEIN", "DISCIPLINE", "ACADEMIC_JOURNAL", "MISC", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_ELEMENT", "ACADEMIC_JOURNAL", "DISCIPLINE"], "gold": {"PROTEIN": ["CCR3", "CC chemokines", "CCL11", "CCL26", "CCL7", "CCL13", "CCL15", "CCL24", "CCL5", "CCL28"], "DISCIPLINE": [], "ACADEMIC_JOURNAL": [], "MISC": ["B and T lymphocyte", "mucosal tissue", "eosinophils", "basophils", "Th1", "Th2", "epithelial cells"], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, DISCIPLINE, ACADEMIC_JOURNAL, MISC, CHEMICAL_ELEMENT\n\nText:\nCCR3 is a receptor for multiple inflammatory / inducible CC chemokines, including CCL11, CCL26, CCL7, CCL13, CCL15, CCL24 and CCL5 that attract eosinophils, and CCL28 that attracts B and T lymphocyte s to mucosal tissue s.ref name = Youn / It is most highly expressed in both eosinophils and basophils, but can also be found in Th1 and Th2 cells and airway epithelial cells.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, DISCIPLINE, ACADEMIC_JOURNAL, MISC, CHEMICAL_ELEMENT\n\nText:\nCCR3 is a receptor for multiple inflammatory / inducible CC chemokines, including CCL11, CCL26, CCL7, CCL13, CCL15, CCL24 and CCL5 that attract eosinophils, and CCL28 that attracts B and T lymphocyte s to mucosal tissue s.ref name = Youn / It is most highly expressed in both eosinophils and basophils, but can also be found in Th1 and Th2 cells and airway epithelial cells.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_130", "text": "On 11 September 2010, Jones performed for an audience of 50,000 at the Help for Heroes charity concert at Twickenham Stadium performing Strange Things Are Happening Every Day and his hit Green, Green Grass of Home.", "requested_types": ["PERSON", "MISC", "AWARD", "ORGANISATION", "EVENT", "LOCATION"], "distractors": ["AWARD"], "gold": {"PERSON": ["Jones"], "MISC": ["Strange Things Are Happening Every Day", "Green, Green Grass of Home"], "AWARD": [], "ORGANISATION": ["Help for Heroes"], "EVENT": ["charity concert"], "LOCATION": ["Twickenham Stadium"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, MISC, AWARD, ORGANISATION, EVENT, LOCATION\n\nText:\nOn 11 September 2010, Jones performed for an audience of 50,000 at the Help for Heroes charity concert at Twickenham Stadium performing Strange Things Are Happening Every Day and his hit Green, Green Grass of Home.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, MISC, AWARD, ORGANISATION, EVENT, LOCATION\n\nText:\nOn 11 September 2010, Jones performed for an audience of 50,000 at the Help for Heroes charity concert at Twickenham Stadium performing Strange Things Are Happening Every Day and his hit Green, Green Grass of Home.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_131", "text": "Mining interests Thomas Ewing and George W. Grayson, then miner Edward Blewett, Judge Hiram G. Bond of Denver and New York City, and Seattle publisher Leigh S. J. Hunt, funded further work and soon a wagon road was built over Barlow Pass to join the Sauk wagon road.", "requested_types": ["ENZYME", "MISC", "PERSON", "LOCATION"], "distractors": ["ENZYME", "MISC"], "gold": {"ENZYME": [], "MISC": [], "PERSON": ["Thomas Ewing", "George W. Grayson", "Edward Blewett", "Hiram G. Bond", "Leigh S. J. Hunt"], "LOCATION": ["Denver", "New York City", "Seattle", "Barlow Pass", "Sauk wagon road"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, MISC, PERSON, LOCATION\n\nText:\nMining interests Thomas Ewing and George W. Grayson, then miner Edward Blewett, Judge Hiram G. Bond of Denver and New York City, and Seattle publisher Leigh S. J. Hunt, funded further work and soon a wagon road was built over Barlow Pass to join the Sauk wagon road.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, MISC, PERSON, LOCATION\n\nText:\nMining interests Thomas Ewing and George W. Grayson, then miner Edward Blewett, Judge Hiram G. Bond of Denver and New York City, and Seattle publisher Leigh S. J. Hunt, funded further work and soon a wagon road was built over Barlow Pass to join the Sauk wagon road.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_132", "text": "Camilla is the 6th trinary asteroid that has been discovered in the asteroid belt, after 87 Sylvia, 45 Eugenia, 216 Kleopatra, 93 Minerva and 130 Elektra.", "requested_types": ["CHEMICAL_ELEMENT", "ORGANISATION", "ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND"], "distractors": ["ORGANISATION", "CHEMICAL_ELEMENT", "CHEMICAL_COMPOUND"], "gold": {"CHEMICAL_ELEMENT": [], "ORGANISATION": [], "ASTRONOMICAL_OBJECT": ["Camilla", "87 Sylvia", "45 Eugenia", "216 Kleopatra", "93 Minerva", "130 Elektra"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, ORGANISATION, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND\n\nText:\nCamilla is the 6th trinary asteroid that has been discovered in the asteroid belt, after 87 Sylvia, 45 Eugenia, 216 Kleopatra, 93 Minerva and 130 Elektra.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, ORGANISATION, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND\n\nText:\nCamilla is the 6th trinary asteroid that has been discovered in the asteroid belt, after 87 Sylvia, 45 Eugenia, 216 Kleopatra, 93 Minerva and 130 Elektra.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_133", "text": "This led to efforts such as Optical Gravitational Lensing Experiment, or OGLE, that have characterized hundreds of such events, including those of OGLE-2016-BLG-1190Lb and OGLE-2016-BLG-1195Lb.", "requested_types": ["CHEMICAL_ELEMENT", "AWARD", "ORGANISATION", "ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND"], "distractors": ["AWARD", "CHEMICAL_ELEMENT", "CHEMICAL_COMPOUND"], "gold": {"CHEMICAL_ELEMENT": [], "AWARD": [], "ORGANISATION": ["Optical Gravitational Lensing Experiment", "OGLE"], "ASTRONOMICAL_OBJECT": ["OGLE-2016-BLG-1190Lb", "OGLE-2016-BLG-1195Lb"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, AWARD, ORGANISATION, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND\n\nText:\nThis led to efforts such as Optical Gravitational Lensing Experiment, or OGLE, that have characterized hundreds of such events, including those of OGLE-2016-BLG-1190Lb and OGLE-2016-BLG-1195Lb.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, AWARD, ORGANISATION, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND\n\nText:\nThis led to efforts such as Optical Gravitational Lensing Experiment, or OGLE, that have characterized hundreds of such events, including those of OGLE-2016-BLG-1190Lb and OGLE-2016-BLG-1195Lb.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_134", "text": "Topics included matter creation and annihilation, the fundamental interaction s, elementary particle s and their currents, hadron ic and lepton ic physics, and the parton model, published in professional peer-reviewed scientific journal s including Nuclear Physics B, Australian Journal of Physics, Nuovo Cimento, and Physical Review D.", "requested_types": ["ACADEMIC_JOURNAL", "MISC", "DISCIPLINE", "EVENT", "AWARD", "COUNTRY"], "distractors": ["AWARD", "EVENT", "COUNTRY"], "gold": {"ACADEMIC_JOURNAL": ["Nuclear Physics B", "Australian Journal of Physics", "Nuovo Cimento", "Physical Review D"], "MISC": ["matter creation", "annihilation", "fundamental interaction", "elementary particle", "hadron ic", "lepton ic", "parton model", "peer-reviewed scientific journal"], "DISCIPLINE": ["physics"], "EVENT": [], "AWARD": [], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, MISC, DISCIPLINE, EVENT, AWARD, COUNTRY\n\nText:\nTopics included matter creation and annihilation, the fundamental interaction s, elementary particle s and their currents, hadron ic and lepton ic physics, and the parton model, published in professional peer-reviewed scientific journal s including Nuclear Physics B, Australian Journal of Physics, Nuovo Cimento, and Physical Review D.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, MISC, DISCIPLINE, EVENT, AWARD, COUNTRY\n\nText:\nTopics included matter creation and annihilation, the fundamental interaction s, elementary particle s and their currents, hadron ic and lepton ic physics, and the parton model, published in professional peer-reviewed scientific journal s including Nuclear Physics B, Australian Journal of Physics, Nuovo Cimento, and Physical Review D.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_135", "text": "In neuron s, gambierol has been reported to induce Ca threshold for the activation of Ras superfamily.", "requested_types": ["MISC", "PROTEIN", "LOCATION", "ENZYME", "CHEMICAL_COMPOUND"], "distractors": ["ENZYME", "LOCATION"], "gold": {"MISC": ["neuron", "Ca threshold"], "PROTEIN": ["Ras superfamily"], "LOCATION": [], "ENZYME": [], "CHEMICAL_COMPOUND": ["gambierol"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, LOCATION, ENZYME, CHEMICAL_COMPOUND\n\nText:\nIn neuron s, gambierol has been reported to induce Ca threshold for the activation of Ras superfamily.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, LOCATION, ENZYME, CHEMICAL_COMPOUND\n\nText:\nIn neuron s, gambierol has been reported to induce Ca threshold for the activation of Ras superfamily.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_136", "text": "It can simulate BOD / DO, Ammonia, Nitrate, Eutrophication, heavy metals and Wetland s.", "requested_types": ["ORGANISATION", "MISC", "PROTEIN", "LOCATION", "CHEMICAL_COMPOUND"], "distractors": ["LOCATION", "ORGANISATION", "PROTEIN"], "gold": {"ORGANISATION": [], "MISC": ["BOD", "DO", "heavy metals", "Wetland"], "PROTEIN": [], "LOCATION": [], "CHEMICAL_COMPOUND": ["Ammonia", "Nitrate", "Eutrophication"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, PROTEIN, LOCATION, CHEMICAL_COMPOUND\n\nText:\nIt can simulate BOD / DO, Ammonia, Nitrate, Eutrophication, heavy metals and Wetland s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, PROTEIN, LOCATION, CHEMICAL_COMPOUND\n\nText:\nIt can simulate BOD / DO, Ammonia, Nitrate, Eutrophication, heavy metals and Wetland s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_137", "text": "The episode's airing - in particular, its shock regeneration - contributed to a public surge of interest, and speculation about Tennant's replacement: actor Robert Carlyle was the bookmaker's favourite and actors James McAvoy, Jason Statham, Alan Davies, and James Nesbitt were less popular predictions.", "requested_types": ["COUNTRY", "AWARD", "PERSON"], "distractors": ["COUNTRY", "AWARD"], "gold": {"COUNTRY": [], "AWARD": [], "PERSON": ["Tennant", "Robert Carlyle", "James McAvoy", "Jason Statham", "Alan Davies", "James Nesbitt"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, AWARD, PERSON\n\nText:\nThe episode's airing - in particular, its shock regeneration - contributed to a public surge of interest, and speculation about Tennant's replacement: actor Robert Carlyle was the bookmaker's favourite and actors James McAvoy, Jason Statham, Alan Davies, and James Nesbitt were less popular predictions.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, AWARD, PERSON\n\nText:\nThe episode's airing - in particular, its shock regeneration - contributed to a public surge of interest, and speculation about Tennant's replacement: actor Robert Carlyle was the bookmaker's favourite and actors James McAvoy, Jason Statham, Alan Davies, and James Nesbitt were less popular predictions.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_138", "text": "He is a member of the American Philosophical Society and an elected Fellow of the American Academy of Arts and Sciences, and in 1985 was a Corresponding Fellow of the British Academy 0.2001.", "requested_types": ["PROTEIN", "AWARD", "ORGANISATION"], "distractors": ["PROTEIN"], "gold": {"PROTEIN": [], "AWARD": ["Fellow of the American Academy of Arts and Sciences", "Fellow of the British Academy"], "ORGANISATION": ["American Philosophical Society"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, AWARD, ORGANISATION\n\nText:\nHe is a member of the American Philosophical Society and an elected Fellow of the American Academy of Arts and Sciences, and in 1985 was a Corresponding Fellow of the British Academy 0.2001.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, AWARD, ORGANISATION\n\nText:\nHe is a member of the American Philosophical Society and an elected Fellow of the American Academy of Arts and Sciences, and in 1985 was a Corresponding Fellow of the British Academy 0.2001.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_139", "text": "Some additional examples include 990 Yerkes, 991 McDonalda, and 992 Swasey around this time; many more minor planets would be discovered at the observatory in the following decades.", "requested_types": ["ENZYME", "ASTRONOMICAL_OBJECT", "ACADEMIC_JOURNAL"], "distractors": ["ACADEMIC_JOURNAL", "ENZYME"], "gold": {"ENZYME": [], "ASTRONOMICAL_OBJECT": ["990 Yerkes", "991 McDonalda", "992 Swasey"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ASTRONOMICAL_OBJECT, ACADEMIC_JOURNAL\n\nText:\nSome additional examples include 990 Yerkes, 991 McDonalda, and 992 Swasey around this time; many more minor planets would be discovered at the observatory in the following decades.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ASTRONOMICAL_OBJECT, ACADEMIC_JOURNAL\n\nText:\nSome additional examples include 990 Yerkes, 991 McDonalda, and 992 Swasey around this time; many more minor planets would be discovered at the observatory in the following decades.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_140", "text": "Yet, unlike vertebrates who use G protein coupled receptors (G protein-coupled receptor), insects express proteins including ORs (olfactory receptors), GRs (gustatory receptors) and IRs (ionotropic receptors) which are all heteromeric ligand-gated ion channels.", "requested_types": ["MISC", "PROTEIN", "THEORY"], "distractors": ["THEORY"], "gold": {"MISC": ["vertebrates", "ORs", "olfactory receptors", "GRs", "gustatory receptors", "IRs", "ionotropic receptors", "heteromeric ligand-gated ion channels"], "PROTEIN": ["G protein coupled receptors", "G protein-coupled receptor"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, THEORY\n\nText:\nYet, unlike vertebrates who use G protein coupled receptors (G protein-coupled receptor), insects express proteins including ORs (olfactory receptors), GRs (gustatory receptors) and IRs (ionotropic receptors) which are all heteromeric ligand-gated ion channels.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, THEORY\n\nText:\nYet, unlike vertebrates who use G protein coupled receptors (G protein-coupled receptor), insects express proteins including ORs (olfactory receptors), GRs (gustatory receptors) and IRs (ionotropic receptors) which are all heteromeric ligand-gated ion channels.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_141", "text": "It works by blocking Type I topoisomerase which results in DNA damage and cell death.", "requested_types": ["ORGANISATION", "PROTEIN", "AWARD", "MISC", "LOCATION"], "distractors": ["ORGANISATION", "LOCATION", "AWARD"], "gold": {"ORGANISATION": [], "PROTEIN": ["Type I topoisomerase"], "AWARD": [], "MISC": ["DNA damage", "cell death"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PROTEIN, AWARD, MISC, LOCATION\n\nText:\nIt works by blocking Type I topoisomerase which results in DNA damage and cell death.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PROTEIN, AWARD, MISC, LOCATION\n\nText:\nIt works by blocking Type I topoisomerase which results in DNA damage and cell death.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_142", "text": "It did not pass through its perihelion until 13 March 1759, the attraction of Jupiter and Saturn having caused a retardation of 618 days.", "requested_types": ["ASTRONOMICAL_OBJECT", "ACADEMIC_JOURNAL", "LOCATION", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_ELEMENT", "ACADEMIC_JOURNAL", "LOCATION"], "gold": {"ASTRONOMICAL_OBJECT": ["Jupiter", "Saturn"], "ACADEMIC_JOURNAL": [], "LOCATION": [], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ACADEMIC_JOURNAL, LOCATION, CHEMICAL_ELEMENT\n\nText:\nIt did not pass through its perihelion until 13 March 1759, the attraction of Jupiter and Saturn having caused a retardation of 618 days.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ACADEMIC_JOURNAL, LOCATION, CHEMICAL_ELEMENT\n\nText:\nIt did not pass through its perihelion until 13 March 1759, the attraction of Jupiter and Saturn having caused a retardation of 618 days.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_143", "text": "The Vital-IT infrastructure is distributed to three sites: University of Lausanne, EPFL and University of Geneva.", "requested_types": ["SCIENTIST", "UNIVERSITY", "ORGANISATION"], "distractors": ["SCIENTIST"], "gold": {"SCIENTIST": [], "UNIVERSITY": ["University of Lausanne", "EPFL", "University of Geneva"], "ORGANISATION": ["Vital-IT"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, UNIVERSITY, ORGANISATION\n\nText:\nThe Vital-IT infrastructure is distributed to three sites: University of Lausanne, EPFL and University of Geneva.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, UNIVERSITY, ORGANISATION\n\nText:\nThe Vital-IT infrastructure is distributed to three sites: University of Lausanne, EPFL and University of Geneva.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_144", "text": "In 2006, his work with WITNESS and his long-standing support of peace and human rights causes was recognised by the Nobel Peace Prize Laureates with the Man of Peace award.", "requested_types": ["ORGANISATION", "MISC", "AWARD"], "distractors": ["MISC"], "gold": {"ORGANISATION": ["WITNESS"], "MISC": [], "AWARD": ["Nobel Peace Prize", "Man of Peace award"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, AWARD\n\nText:\nIn 2006, his work with WITNESS and his long-standing support of peace and human rights causes was recognised by the Nobel Peace Prize Laureates with the Man of Peace award.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, AWARD\n\nText:\nIn 2006, his work with WITNESS and his long-standing support of peace and human rights causes was recognised by the Nobel Peace Prize Laureates with the Man of Peace award.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_145", "text": "Three of the newly confirmed exoplanets were found to orbit within habitable zones of their related star s: two of the three, Kepler-438b and Kepler-442b, are near-Earth-size and likely rocky; the third, Kepler-440b, is a super-Earth.", "requested_types": ["MISC", "AWARD", "ASTRONOMICAL_OBJECT", "ENZYME"], "distractors": ["ENZYME", "AWARD"], "gold": {"MISC": ["exoplanets", "super-Earth"], "AWARD": [], "ASTRONOMICAL_OBJECT": ["Kepler-438b", "Kepler-442b", "Kepler-440b"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, AWARD, ASTRONOMICAL_OBJECT, ENZYME\n\nText:\nThree of the newly confirmed exoplanets were found to orbit within habitable zones of their related star s: two of the three, Kepler-438b and Kepler-442b, are near-Earth-size and likely rocky; the third, Kepler-440b, is a super-Earth.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, AWARD, ASTRONOMICAL_OBJECT, ENZYME\n\nText:\nThree of the newly confirmed exoplanets were found to orbit within habitable zones of their related star s: two of the three, Kepler-438b and Kepler-442b, are near-Earth-size and likely rocky; the third, Kepler-440b, is a super-Earth.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_146", "text": "Alessandra Becatti (born 30 April 1965) is an Italy female retired heptathlete, which participated at the 1987 World Championships in Athletics.", "requested_types": ["EVENT", "PERSON", "CHEMICAL_ELEMENT", "COUNTRY"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"EVENT": ["1987 World Championships in Athletics"], "PERSON": ["Alessandra Becatti"], "CHEMICAL_ELEMENT": [], "COUNTRY": ["Italy"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, PERSON, CHEMICAL_ELEMENT, COUNTRY\n\nText:\nAlessandra Becatti (born 30 April 1965) is an Italy female retired heptathlete, which participated at the 1987 World Championships in Athletics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, PERSON, CHEMICAL_ELEMENT, COUNTRY\n\nText:\nAlessandra Becatti (born 30 April 1965) is an Italy female retired heptathlete, which participated at the 1987 World Championships in Athletics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_147", "text": "In order to reset the protein to its original, basic state, it needs ATP synthase.", "requested_types": ["SCIENTIST", "ENZYME", "PERSON"], "distractors": ["SCIENTIST", "PERSON"], "gold": {"SCIENTIST": [], "ENZYME": ["ATP synthase"], "PERSON": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ENZYME, PERSON\n\nText:\nIn order to reset the protein to its original, basic state, it needs ATP synthase.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ENZYME, PERSON\n\nText:\nIn order to reset the protein to its original, basic state, it needs ATP synthase.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_148", "text": "In addition to the lighthouse, there are four other sites in Point Loma listed on the National Register of Historic Places: Cabrillo National Monument, the Marine Corps Recruit Depot San Diego, Naval Training Center San Diego, and Rosecroft.", "requested_types": ["THEORY", "LOCATION", "MISC"], "distractors": ["THEORY"], "gold": {"THEORY": [], "LOCATION": ["Point Loma", "Cabrillo National Monument", "Marine Corps Recruit Depot San Diego", "Naval Training Center San Diego", "Rosecroft"], "MISC": ["National Register of Historic Places"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, LOCATION, MISC\n\nText:\nIn addition to the lighthouse, there are four other sites in Point Loma listed on the National Register of Historic Places: Cabrillo National Monument, the Marine Corps Recruit Depot San Diego, Naval Training Center San Diego, and Rosecroft.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, LOCATION, MISC\n\nText:\nIn addition to the lighthouse, there are four other sites in Point Loma listed on the National Register of Historic Places: Cabrillo National Monument, the Marine Corps Recruit Depot San Diego, Naval Training Center San Diego, and Rosecroft.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_149", "text": "In November 2018 a group of Chinese scientists published research in the journal Molecular Therapy detailing their use of CRISPR technology to correct a single mistaken amino acid successfully in 16 out of 18 attempts in a human embryo.", "requested_types": ["MISC", "ACADEMIC_JOURNAL", "LOCATION", "CHEMICAL_COMPOUND"], "distractors": ["LOCATION"], "gold": {"MISC": ["Chinese", "CRISPR technology"], "ACADEMIC_JOURNAL": ["Molecular Therapy"], "LOCATION": [], "CHEMICAL_COMPOUND": ["amino acid"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ACADEMIC_JOURNAL, LOCATION, CHEMICAL_COMPOUND\n\nText:\nIn November 2018 a group of Chinese scientists published research in the journal Molecular Therapy detailing their use of CRISPR technology to correct a single mistaken amino acid successfully in 16 out of 18 attempts in a human embryo.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ACADEMIC_JOURNAL, LOCATION, CHEMICAL_COMPOUND\n\nText:\nIn November 2018 a group of Chinese scientists published research in the journal Molecular Therapy detailing their use of CRISPR technology to correct a single mistaken amino acid successfully in 16 out of 18 attempts in a human embryo.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_150", "text": "With the first group interacts one plaque protein without PDZ domain, called cingulin, which plays a key role in the cell adhesion.", "requested_types": ["ORGANISATION", "PROTEIN", "MISC"], "distractors": ["ORGANISATION"], "gold": {"ORGANISATION": [], "PROTEIN": ["plaque protein", "cingulin"], "MISC": ["PDZ domain", "cell adhesion"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PROTEIN, MISC\n\nText:\nWith the first group interacts one plaque protein without PDZ domain, called cingulin, which plays a key role in the cell adhesion.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PROTEIN, MISC\n\nText:\nWith the first group interacts one plaque protein without PDZ domain, called cingulin, which plays a key role in the cell adhesion.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_151", "text": "The story instead focuses on the Doctor's companion, Donna Noble (Catherine Tate) and her encounters with former companion Rose Tyler (Billie Piper).", "requested_types": ["THEORY", "ACADEMIC_JOURNAL", "AWARD", "PERSON"], "distractors": ["THEORY", "ACADEMIC_JOURNAL", "AWARD"], "gold": {"THEORY": [], "ACADEMIC_JOURNAL": [], "AWARD": [], "PERSON": ["Donna Noble", "Catherine Tate", "Rose Tyler", "Billie Piper"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ACADEMIC_JOURNAL, AWARD, PERSON\n\nText:\nThe story instead focuses on the Doctor's companion, Donna Noble (Catherine Tate) and her encounters with former companion Rose Tyler (Billie Piper).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ACADEMIC_JOURNAL, AWARD, PERSON\n\nText:\nThe story instead focuses on the Doctor's companion, Donna Noble (Catherine Tate) and her encounters with former companion Rose Tyler (Billie Piper).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_152", "text": "In addition, he finished fourth at the 2014 European Championships and won the silver medal at the 2013 Summer Universiade.", "requested_types": ["EVENT", "PROTEIN", "UNIVERSITY", "CHEMICAL_COMPOUND"], "distractors": ["PROTEIN", "CHEMICAL_COMPOUND", "UNIVERSITY"], "gold": {"EVENT": ["2014 European Championships", "2013 Summer Universiade"], "PROTEIN": [], "UNIVERSITY": [], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, PROTEIN, UNIVERSITY, CHEMICAL_COMPOUND\n\nText:\nIn addition, he finished fourth at the 2014 European Championships and won the silver medal at the 2013 Summer Universiade.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, PROTEIN, UNIVERSITY, CHEMICAL_COMPOUND\n\nText:\nIn addition, he finished fourth at the 2014 European Championships and won the silver medal at the 2013 Summer Universiade.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_153", "text": "As early as 1807 the degree of LL.D. was conferred upon Brewster by Marischal College, Aberdeen; in 1815 he was elected a Fellow of the Royal Society, and received the Copley Medal; in 1818 he received the Rumford Medal of the society; and in 1816 the French Institute awarded him one-half of the prize of three thousand francs for the two most important discoveries in physical science made in Europe during the two preceding years.", "requested_types": ["CHEMICAL_COMPOUND", "CHEMICAL_ELEMENT", "DISCIPLINE", "LOCATION", "ORGANISATION", "AWARD", "PERSON", "UNIVERSITY", "COUNTRY", "MISC"], "distractors": ["CHEMICAL_ELEMENT", "CHEMICAL_COMPOUND", "COUNTRY"], "gold": {"CHEMICAL_COMPOUND": [], "CHEMICAL_ELEMENT": [], "DISCIPLINE": ["physical science"], "LOCATION": ["Aberdeen", "Europe"], "ORGANISATION": ["French Institute"], "AWARD": ["Fellow of the Royal Society", "Copley Medal", "Rumford Medal"], "PERSON": ["Brewster"], "UNIVERSITY": ["Marischal College"], "COUNTRY": [], "MISC": ["LL.D."]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, CHEMICAL_ELEMENT, DISCIPLINE, LOCATION, ORGANISATION, AWARD, PERSON, UNIVERSITY, COUNTRY, MISC\n\nText:\nAs early as 1807 the degree of LL.D. was conferred upon Brewster by Marischal College, Aberdeen; in 1815 he was elected a Fellow of the Royal Society, and received the Copley Medal; in 1818 he received the Rumford Medal of the society; and in 1816 the French Institute awarded him one-half of the prize of three thousand francs for the two most important discoveries in physical science made in Europe during the two preceding years.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, CHEMICAL_ELEMENT, DISCIPLINE, LOCATION, ORGANISATION, AWARD, PERSON, UNIVERSITY, COUNTRY, MISC\n\nText:\nAs early as 1807 the degree of LL.D. was conferred upon Brewster by Marischal College, Aberdeen; in 1815 he was elected a Fellow of the Royal Society, and received the Copley Medal; in 1818 he received the Rumford Medal of the society; and in 1816 the French Institute awarded him one-half of the prize of three thousand francs for the two most important discoveries in physical science made in Europe during the two preceding years.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_154", "text": "There are several descriptions and discussions about a philosophical theory of the water cycle by Vitruvius, Leonardo da Vinci and Bernard Palissy.", "requested_types": ["THEORY", "PERSON", "DISCIPLINE", "MISC"], "distractors": ["THEORY"], "gold": {"THEORY": [], "PERSON": ["Vitruvius", "Leonardo da Vinci", "Bernard Palissy"], "DISCIPLINE": ["philosophical"], "MISC": ["water cycle"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, PERSON, DISCIPLINE, MISC\n\nText:\nThere are several descriptions and discussions about a philosophical theory of the water cycle by Vitruvius, Leonardo da Vinci and Bernard Palissy.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, PERSON, DISCIPLINE, MISC\n\nText:\nThere are several descriptions and discussions about a philosophical theory of the water cycle by Vitruvius, Leonardo da Vinci and Bernard Palissy.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_155", "text": "Before receiving his Ph.D. for atomic models of alkali metals from Massachusetts Institute of Technology in 1924, he spent the 1922-23 academic year as a Fellow of The American-Scandinavian Foundation at the University of Copenhagen under Niels Bohr and Hans Kramers.", "requested_types": ["THEORY", "AWARD", "SCIENTIST", "ORGANISATION", "UNIVERSITY", "MISC"], "distractors": ["ORGANISATION", "THEORY"], "gold": {"THEORY": [], "AWARD": ["Fellow of The American-Scandinavian Foundation"], "SCIENTIST": ["Niels Bohr", "Hans Kramers"], "ORGANISATION": [], "UNIVERSITY": ["Massachusetts Institute of Technology", "University of Copenhagen"], "MISC": ["alkali metals"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, AWARD, SCIENTIST, ORGANISATION, UNIVERSITY, MISC\n\nText:\nBefore receiving his Ph.D. for atomic models of alkali metals from Massachusetts Institute of Technology in 1924, he spent the 1922-23 academic year as a Fellow of The American-Scandinavian Foundation at the University of Copenhagen under Niels Bohr and Hans Kramers.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, AWARD, SCIENTIST, ORGANISATION, UNIVERSITY, MISC\n\nText:\nBefore receiving his Ph.D. for atomic models of alkali metals from Massachusetts Institute of Technology in 1924, he spent the 1922-23 academic year as a Fellow of The American-Scandinavian Foundation at the University of Copenhagen under Niels Bohr and Hans Kramers.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_156", "text": "Encke (1850) has further symbols for 5 Astraea, 6 Hebe, 7 Iris, 8 Flora and 9 Metis.", "requested_types": ["ASTRONOMICAL_OBJECT", "THEORY", "MISC"], "distractors": ["MISC", "THEORY"], "gold": {"ASTRONOMICAL_OBJECT": ["Encke", "5 Astraea", "6 Hebe", "7 Iris", "8 Flora", "9 Metis"], "THEORY": [], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, THEORY, MISC\n\nText:\nEncke (1850) has further symbols for 5 Astraea, 6 Hebe, 7 Iris, 8 Flora and 9 Metis.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, THEORY, MISC\n\nText:\nEncke (1850) has further symbols for 5 Astraea, 6 Hebe, 7 Iris, 8 Flora and 9 Metis.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_157", "text": "ABC Family Worldwide Inc (ABC Family Channel), Regent University, the American Center for Law & Justice (ACLJ), the Founders Inn and Conference Center, the Christian Coalition of America, an L-1011 Flying Hospital, Operation Blessing International, and CBN Asia.", "requested_types": ["MISC", "UNIVERSITY", "SCIENTIST", "ORGANISATION", "ASTRONOMICAL_OBJECT"], "distractors": ["ASTRONOMICAL_OBJECT", "SCIENTIST"], "gold": {"MISC": ["ABC Family Channel"], "UNIVERSITY": ["Regent University"], "SCIENTIST": [], "ORGANISATION": ["ABC Family Worldwide Inc", "American Center for Law & Justice", "ACLJ", "Founders Inn and Conference Center", "Christian Coalition of America", "L-1011 Flying Hospital", "Operation Blessing International", "CBN Asia"], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, UNIVERSITY, SCIENTIST, ORGANISATION, ASTRONOMICAL_OBJECT\n\nText:\nABC Family Worldwide Inc (ABC Family Channel), Regent University, the American Center for Law & Justice (ACLJ), the Founders Inn and Conference Center, the Christian Coalition of America, an L-1011 Flying Hospital, Operation Blessing International, and CBN Asia.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, UNIVERSITY, SCIENTIST, ORGANISATION, ASTRONOMICAL_OBJECT\n\nText:\nABC Family Worldwide Inc (ABC Family Channel), Regent University, the American Center for Law & Justice (ACLJ), the Founders Inn and Conference Center, the Christian Coalition of America, an L-1011 Flying Hospital, Operation Blessing International, and CBN Asia.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_158", "text": "The most common herbicide used was Herbicide Orange, more commonly referred to as Agent Orange: a fifty-fifty mixture of two herbicides 2,4-Dichlorophenoxyacetic acid (2,4-dichlorophenoxyacetic acid) and 2,4,5-Trichlorophenoxyacetic acid (2,4,5-trichlorophenoxyacetic acid) manufactured for the U.S. Department of Defense primarily by Monsanto Corporation and Dow Chemical.", "requested_types": ["CHEMICAL_COMPOUND", "ORGANISATION", "CHEMICAL_ELEMENT", "LOCATION"], "distractors": ["CHEMICAL_ELEMENT", "LOCATION"], "gold": {"CHEMICAL_COMPOUND": ["Herbicide Orange", "Agent Orange", "2,4-Dichlorophenoxyacetic acid", "2,4-dichlorophenoxyacetic acid", "2,4,5-Trichlorophenoxyacetic acid", "2,4,5-trichlorophenoxyacetic acid"], "ORGANISATION": ["U.S. Department of Defense", "Monsanto Corporation", "Dow Chemical"], "CHEMICAL_ELEMENT": [], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ORGANISATION, CHEMICAL_ELEMENT, LOCATION\n\nText:\nThe most common herbicide used was Herbicide Orange, more commonly referred to as Agent Orange: a fifty-fifty mixture of two herbicides 2,4-Dichlorophenoxyacetic acid (2,4-dichlorophenoxyacetic acid) and 2,4,5-Trichlorophenoxyacetic acid (2,4,5-trichlorophenoxyacetic acid) manufactured for the U.S. Department of Defense primarily by Monsanto Corporation and Dow Chemical.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ORGANISATION, CHEMICAL_ELEMENT, LOCATION\n\nText:\nThe most common herbicide used was Herbicide Orange, more commonly referred to as Agent Orange: a fifty-fifty mixture of two herbicides 2,4-Dichlorophenoxyacetic acid (2,4-dichlorophenoxyacetic acid) and 2,4,5-Trichlorophenoxyacetic acid (2,4,5-trichlorophenoxyacetic acid) manufactured for the U.S. Department of Defense primarily by Monsanto Corporation and Dow Chemical.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_159", "text": "During his career he held academic posts at the universities of University of Manchester (1951-6), Durham University (1956-92), the University of Ceylon and the University of Hong Kong, and was head of department at Durham where he remains an emeritus professor.", "requested_types": ["CHEMICAL_ELEMENT", "UNIVERSITY"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_ELEMENT": [], "UNIVERSITY": ["University of Manchester", "Durham University", "University of Ceylon", "University of Hong Kong", "Durham"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, UNIVERSITY\n\nText:\nDuring his career he held academic posts at the universities of University of Manchester (1951-6), Durham University (1956-92), the University of Ceylon and the University of Hong Kong, and was head of department at Durham where he remains an emeritus professor.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, UNIVERSITY\n\nText:\nDuring his career he held academic posts at the universities of University of Manchester (1951-6), Durham University (1956-92), the University of Ceylon and the University of Hong Kong, and was head of department at Durham where he remains an emeritus professor.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_160", "text": "Throughout its run, the executive producers of the series were Damon Lindelof, J. J. Abrams, Bryan Burk, Carlton Cuse, Jack Bender, Jeff Pinkner, Edward Kitsis, Adam Horowitz, Jean Higgins, and Elizabeth Sarnoff, with Lindelof and Cuse serving as showrunner s.", "requested_types": ["LOCATION", "PERSON", "PROTEIN"], "distractors": ["LOCATION", "PROTEIN"], "gold": {"LOCATION": [], "PERSON": ["Damon Lindelof", "J. J. Abrams", "Bryan Burk", "Carlton Cuse", "Jack Bender", "Jeff Pinkner", "Edward Kitsis", "Adam Horowitz", "Jean Higgins", "Elizabeth Sarnoff", "Lindelof"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, PERSON, PROTEIN\n\nText:\nThroughout its run, the executive producers of the series were Damon Lindelof, J. J. Abrams, Bryan Burk, Carlton Cuse, Jack Bender, Jeff Pinkner, Edward Kitsis, Adam Horowitz, Jean Higgins, and Elizabeth Sarnoff, with Lindelof and Cuse serving as showrunner s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, PERSON, PROTEIN\n\nText:\nThroughout its run, the executive producers of the series were Damon Lindelof, J. J. Abrams, Bryan Burk, Carlton Cuse, Jack Bender, Jeff Pinkner, Edward Kitsis, Adam Horowitz, Jean Higgins, and Elizabeth Sarnoff, with Lindelof and Cuse serving as showrunner s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_161", "text": "In 1896, he translated, from German, Heimat, a play in four acts by Hermann Sudermann, renamed Magda and played by Henry Stephenson and Charles Waldron in a Broadway theatre production in New York City, New York.", "requested_types": ["PERSON", "THEORY", "DISCIPLINE", "MISC", "LOCATION"], "distractors": ["DISCIPLINE", "THEORY"], "gold": {"PERSON": ["Hermann Sudermann", "Magda", "Henry Stephenson", "Charles Waldron"], "THEORY": [], "DISCIPLINE": [], "MISC": ["German", "Broadway theatre"], "LOCATION": ["New York City", "New York"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, THEORY, DISCIPLINE, MISC, LOCATION\n\nText:\nIn 1896, he translated, from German, Heimat, a play in four acts by Hermann Sudermann, renamed Magda and played by Henry Stephenson and Charles Waldron in a Broadway theatre production in New York City, New York.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, THEORY, DISCIPLINE, MISC, LOCATION\n\nText:\nIn 1896, he translated, from German, Heimat, a play in four acts by Hermann Sudermann, renamed Magda and played by Henry Stephenson and Charles Waldron in a Broadway theatre production in New York City, New York.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_162", "text": "It was discovered on 24 September 1960, by Ingrid van Houten-Groeneveld and Cornelis Johannes van Houten at Leiden, on photographic plates taken by Tom Gehrels at Palomar Observatory in the United States.", "requested_types": ["AWARD", "SCIENTIST", "PERSON", "LOCATION", "COUNTRY"], "distractors": ["AWARD"], "gold": {"AWARD": [], "SCIENTIST": ["Ingrid van Houten-Groeneveld", "Cornelis Johannes van Houten"], "PERSON": ["Tom Gehrels"], "LOCATION": ["Leiden", "Palomar Observatory"], "COUNTRY": ["United States"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, SCIENTIST, PERSON, LOCATION, COUNTRY\n\nText:\nIt was discovered on 24 September 1960, by Ingrid van Houten-Groeneveld and Cornelis Johannes van Houten at Leiden, on photographic plates taken by Tom Gehrels at Palomar Observatory in the United States.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, SCIENTIST, PERSON, LOCATION, COUNTRY\n\nText:\nIt was discovered on 24 September 1960, by Ingrid van Houten-Groeneveld and Cornelis Johannes van Houten at Leiden, on photographic plates taken by Tom Gehrels at Palomar Observatory in the United States.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_163", "text": "On the other three gas giant s (Saturn, Uranus and Neptune) eclipses only occur at certain periods during the planet's orbit, due to their higher inclination between the orbits of the moon and the orbital plane of the planet.", "requested_types": ["AWARD", "ASTRONOMICAL_OBJECT", "LOCATION", "ACADEMIC_JOURNAL", "MISC"], "distractors": ["ACADEMIC_JOURNAL", "LOCATION", "AWARD"], "gold": {"AWARD": [], "ASTRONOMICAL_OBJECT": ["Saturn", "Uranus", "Neptune"], "LOCATION": [], "ACADEMIC_JOURNAL": [], "MISC": ["gas giant", "moon"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ASTRONOMICAL_OBJECT, LOCATION, ACADEMIC_JOURNAL, MISC\n\nText:\nOn the other three gas giant s (Saturn, Uranus and Neptune) eclipses only occur at certain periods during the planet's orbit, due to their higher inclination between the orbits of the moon and the orbital plane of the planet.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ASTRONOMICAL_OBJECT, LOCATION, ACADEMIC_JOURNAL, MISC\n\nText:\nOn the other three gas giant s (Saturn, Uranus and Neptune) eclipses only occur at certain periods during the planet's orbit, due to their higher inclination between the orbits of the moon and the orbital plane of the planet.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_164", "text": "It stars Vincent Gallo and Maria de Medeiros, and was directed by Spanish filmmaker and actress María Lidón (credited in the English version of the movie as Luna), with screenplay by Spanish science fiction author Juan Miguel Aguilera.", "requested_types": ["MISC", "PERSON", "UNIVERSITY"], "distractors": ["UNIVERSITY"], "gold": {"MISC": ["Spanish", "English", "Luna"], "PERSON": ["Vincent Gallo", "Maria de Medeiros", "María Lidón", "Juan Miguel Aguilera"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PERSON, UNIVERSITY\n\nText:\nIt stars Vincent Gallo and Maria de Medeiros, and was directed by Spanish filmmaker and actress María Lidón (credited in the English version of the movie as Luna), with screenplay by Spanish science fiction author Juan Miguel Aguilera.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PERSON, UNIVERSITY\n\nText:\nIt stars Vincent Gallo and Maria de Medeiros, and was directed by Spanish filmmaker and actress María Lidón (credited in the English version of the movie as Luna), with screenplay by Spanish science fiction author Juan Miguel Aguilera.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_165", "text": "Occultations have also been used to estimate the diameter of trans-Neptunian object s such as, 28978 Ixion, and 20000 Varuna.", "requested_types": ["MISC", "ENZYME", "ASTRONOMICAL_OBJECT", "LOCATION", "DISCIPLINE"], "distractors": ["LOCATION", "DISCIPLINE", "ENZYME"], "gold": {"MISC": ["Occultations"], "ENZYME": [], "ASTRONOMICAL_OBJECT": ["trans-Neptunian object", "28978 Ixion", "20000 Varuna"], "LOCATION": [], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ENZYME, ASTRONOMICAL_OBJECT, LOCATION, DISCIPLINE\n\nText:\nOccultations have also been used to estimate the diameter of trans-Neptunian object s such as, 28978 Ixion, and 20000 Varuna.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ENZYME, ASTRONOMICAL_OBJECT, LOCATION, DISCIPLINE\n\nText:\nOccultations have also been used to estimate the diameter of trans-Neptunian object s such as, 28978 Ixion, and 20000 Varuna.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_166", "text": "As demonstrated by Temin and Baltimore, who shared the Nobel Prize in Physiology or Medicine with Dulbecco, the transfer of viral genes to the cell is mediated by an enzyme called reverse transcriptase (or, more precisely, RNA-dependent DNA polymerase), which replicates the viral genome (in this case made of RNA) into DNA, which is later incorporated in the host genome.", "requested_types": ["MISC", "ENZYME", "ASTRONOMICAL_OBJECT", "AWARD", "SCIENTIST"], "distractors": ["ASTRONOMICAL_OBJECT"], "gold": {"MISC": ["RNA", "DNA"], "ENZYME": ["reverse transcriptase", "RNA-dependent DNA polymerase"], "ASTRONOMICAL_OBJECT": [], "AWARD": ["Nobel Prize in Physiology or Medicine"], "SCIENTIST": ["Temin", "Baltimore", "Dulbecco"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ENZYME, ASTRONOMICAL_OBJECT, AWARD, SCIENTIST\n\nText:\nAs demonstrated by Temin and Baltimore, who shared the Nobel Prize in Physiology or Medicine with Dulbecco, the transfer of viral genes to the cell is mediated by an enzyme called reverse transcriptase (or, more precisely, RNA-dependent DNA polymerase), which replicates the viral genome (in this case made of RNA) into DNA, which is later incorporated in the host genome.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ENZYME, ASTRONOMICAL_OBJECT, AWARD, SCIENTIST\n\nText:\nAs demonstrated by Temin and Baltimore, who shared the Nobel Prize in Physiology or Medicine with Dulbecco, the transfer of viral genes to the cell is mediated by an enzyme called reverse transcriptase (or, more precisely, RNA-dependent DNA polymerase), which replicates the viral genome (in this case made of RNA) into DNA, which is later incorporated in the host genome.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_167", "text": "In June 2011, during the Second CoRoT Symposium, the probe added ten new objects to the Exoplanet catalogue: CoRoT-16b, CoRoT-17b, CoRoT-18b, CoRoT-19b, CoRoT-20b, CoRoT-21b, CoRoT-22b, CoRoT-23b, CoRoT-24b, CoRoT-24c.", "requested_types": ["EVENT", "CHEMICAL_ELEMENT", "ASTRONOMICAL_OBJECT", "MISC"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"EVENT": ["Second CoRoT Symposium"], "CHEMICAL_ELEMENT": [], "ASTRONOMICAL_OBJECT": ["CoRoT-16b", "CoRoT-17b", "CoRoT-18b", "CoRoT-19b", "CoRoT-20b", "CoRoT-21b", "CoRoT-22b", "CoRoT-23b", "CoRoT-24b", "CoRoT-24c"], "MISC": ["Exoplanet catalogue"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, CHEMICAL_ELEMENT, ASTRONOMICAL_OBJECT, MISC\n\nText:\nIn June 2011, during the Second CoRoT Symposium, the probe added ten new objects to the Exoplanet catalogue: CoRoT-16b, CoRoT-17b, CoRoT-18b, CoRoT-19b, CoRoT-20b, CoRoT-21b, CoRoT-22b, CoRoT-23b, CoRoT-24b, CoRoT-24c.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, CHEMICAL_ELEMENT, ASTRONOMICAL_OBJECT, MISC\n\nText:\nIn June 2011, during the Second CoRoT Symposium, the probe added ten new objects to the Exoplanet catalogue: CoRoT-16b, CoRoT-17b, CoRoT-18b, CoRoT-19b, CoRoT-20b, CoRoT-21b, CoRoT-22b, CoRoT-23b, CoRoT-24b, CoRoT-24c.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_168", "text": "When he was about eighteen he went to the University of Copenhagen and afterwards studied at University of Rostock and Wittenberg.", "requested_types": ["UNIVERSITY", "ASTRONOMICAL_OBJECT"], "distractors": ["ASTRONOMICAL_OBJECT"], "gold": {"UNIVERSITY": ["University of Copenhagen", "University of Rostock and Wittenberg"], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ASTRONOMICAL_OBJECT\n\nText:\nWhen he was about eighteen he went to the University of Copenhagen and afterwards studied at University of Rostock and Wittenberg.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ASTRONOMICAL_OBJECT\n\nText:\nWhen he was about eighteen he went to the University of Copenhagen and afterwards studied at University of Rostock and Wittenberg.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_169", "text": "Together with Roger Guillemin he described the neurohormone GnRH that controls Follicle-stimulating hormone and Luteinizing hormone.", "requested_types": ["SCIENTIST", "PROTEIN", "ENZYME", "MISC"], "distractors": ["ENZYME"], "gold": {"SCIENTIST": ["Roger Guillemin"], "PROTEIN": ["Follicle-stimulating hormone", "Luteinizing hormone"], "ENZYME": [], "MISC": ["neurohormone", "GnRH"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, PROTEIN, ENZYME, MISC\n\nText:\nTogether with Roger Guillemin he described the neurohormone GnRH that controls Follicle-stimulating hormone and Luteinizing hormone.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, PROTEIN, ENZYME, MISC\n\nText:\nTogether with Roger Guillemin he described the neurohormone GnRH that controls Follicle-stimulating hormone and Luteinizing hormone.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_170", "text": "In December 2015, it was recognized as one of four new elements by the Joint Working Party of international scientific bodies International Union of Pure and Applied Chemistry and International Union of Pure and Applied Physics.", "requested_types": ["ORGANISATION", "ASTRONOMICAL_OBJECT"], "distractors": ["ASTRONOMICAL_OBJECT"], "gold": {"ORGANISATION": ["Joint Working Party", "International Union of Pure and Applied Chemistry", "International Union of Pure and Applied Physics"], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ASTRONOMICAL_OBJECT\n\nText:\nIn December 2015, it was recognized as one of four new elements by the Joint Working Party of international scientific bodies International Union of Pure and Applied Chemistry and International Union of Pure and Applied Physics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ASTRONOMICAL_OBJECT\n\nText:\nIn December 2015, it was recognized as one of four new elements by the Joint Working Party of international scientific bodies International Union of Pure and Applied Chemistry and International Union of Pure and Applied Physics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_171", "text": "Since 1993 the Milutin Milankovic Medal has been awarded by the European Geophysical Society (called the European Geosciences Union since 2003) for contributions in the", "requested_types": ["AWARD", "DISCIPLINE", "ORGANISATION", "LOCATION"], "distractors": ["DISCIPLINE", "LOCATION"], "gold": {"AWARD": ["Milutin Milankovic Medal"], "DISCIPLINE": [], "ORGANISATION": ["European Geophysical Society", "European Geosciences Union"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, DISCIPLINE, ORGANISATION, LOCATION\n\nText:\nSince 1993 the Milutin Milankovic Medal has been awarded by the European Geophysical Society (called the European Geosciences Union since 2003) for contributions in the\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, DISCIPLINE, ORGANISATION, LOCATION\n\nText:\nSince 1993 the Milutin Milankovic Medal has been awarded by the European Geophysical Society (called the European Geosciences Union since 2003) for contributions in the\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_172", "text": "The invitees included Walther Bothe, Siegfried Flügge, Hans Geiger, Otto Hahn, Paul Harteck, Gerhard Hoffmann, Josef Mattauch, and Georg Stetter.", "requested_types": ["ENZYME", "ASTRONOMICAL_OBJECT", "DISCIPLINE", "SCIENTIST"], "distractors": ["DISCIPLINE", "ENZYME", "ASTRONOMICAL_OBJECT"], "gold": {"ENZYME": [], "ASTRONOMICAL_OBJECT": [], "DISCIPLINE": [], "SCIENTIST": ["Walther Bothe", "Siegfried Flügge", "Hans Geiger", "Otto Hahn", "Paul Harteck", "Gerhard Hoffmann", "Josef Mattauch", "Georg Stetter"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ASTRONOMICAL_OBJECT, DISCIPLINE, SCIENTIST\n\nText:\nThe invitees included Walther Bothe, Siegfried Flügge, Hans Geiger, Otto Hahn, Paul Harteck, Gerhard Hoffmann, Josef Mattauch, and Georg Stetter.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ASTRONOMICAL_OBJECT, DISCIPLINE, SCIENTIST\n\nText:\nThe invitees included Walther Bothe, Siegfried Flügge, Hans Geiger, Otto Hahn, Paul Harteck, Gerhard Hoffmann, Josef Mattauch, and Georg Stetter.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_173", "text": "The study showed that CRISPR is could effectively be used as a gene-editing tool in human 2PN zygotes, which could lead potentially pregnancy viable if implanted.", "requested_types": ["LOCATION", "MISC"], "distractors": ["LOCATION"], "gold": {"LOCATION": [], "MISC": ["CRISPR", "human 2PN zygotes"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, MISC\n\nText:\nThe study showed that CRISPR is could effectively be used as a gene-editing tool in human 2PN zygotes, which could lead potentially pregnancy viable if implanted.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, MISC\n\nText:\nThe study showed that CRISPR is could effectively be used as a gene-editing tool in human 2PN zygotes, which could lead potentially pregnancy viable if implanted.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_174", "text": "The Editors-in-Chief of Neuro-Ophthalmology are Gordon Plant (National Hospital for Neurology and Neurosurgery, London, UK) and Walter Jay (Loyola University Medical Center).", "requested_types": ["PERSON", "ACADEMIC_JOURNAL", "LOCATION", "ORGANISATION", "COUNTRY", "SCIENTIST"], "distractors": ["PERSON"], "gold": {"PERSON": [], "ACADEMIC_JOURNAL": ["Neuro-Ophthalmology"], "LOCATION": ["London"], "ORGANISATION": ["National Hospital for Neurology and Neurosurgery", "Loyola University Medical Center"], "COUNTRY": ["UK"], "SCIENTIST": ["Gordon Plant", "Walter Jay"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ACADEMIC_JOURNAL, LOCATION, ORGANISATION, COUNTRY, SCIENTIST\n\nText:\nThe Editors-in-Chief of Neuro-Ophthalmology are Gordon Plant (National Hospital for Neurology and Neurosurgery, London, UK) and Walter Jay (Loyola University Medical Center).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ACADEMIC_JOURNAL, LOCATION, ORGANISATION, COUNTRY, SCIENTIST\n\nText:\nThe Editors-in-Chief of Neuro-Ophthalmology are Gordon Plant (National Hospital for Neurology and Neurosurgery, London, UK) and Walter Jay (Loyola University Medical Center).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_175", "text": "MACROH2A deposition has been suggested to be regulated by the Cullin - SPOP - RBX1 ligase complex and is actively involved in stable X inactivation, likely through the formation of an additional layer of epigenetic silencing.", "requested_types": ["ENZYME", "MISC", "PROTEIN"], "distractors": ["ENZYME"], "gold": {"ENZYME": [], "MISC": ["MACROH2A deposition", "X inactivation", "epigenetic silencing"], "PROTEIN": ["Cullin - SPOP - RBX1 ligase complex"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, MISC, PROTEIN\n\nText:\nMACROH2A deposition has been suggested to be regulated by the Cullin - SPOP - RBX1 ligase complex and is actively involved in stable X inactivation, likely through the formation of an additional layer of epigenetic silencing.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, MISC, PROTEIN\n\nText:\nMACROH2A deposition has been suggested to be regulated by the Cullin - SPOP - RBX1 ligase complex and is actively involved in stable X inactivation, likely through the formation of an additional layer of epigenetic silencing.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_176", "text": "In addition, three new actors joined the regular cast: Elizabeth Mitchell, as fertility doctor and Other Juliet Burke; and Kiele Sanchez and Rodrigo Santoro as background survivor couple Nikki Fernandez and Paulo.", "requested_types": ["PERSON", "AWARD"], "distractors": ["AWARD"], "gold": {"PERSON": ["Elizabeth Mitchell", "Juliet Burke", "Kiele Sanchez", "Rodrigo Santoro", "Nikki Fernandez", "Paulo"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, AWARD\n\nText:\nIn addition, three new actors joined the regular cast: Elizabeth Mitchell, as fertility doctor and Other Juliet Burke; and Kiele Sanchez and Rodrigo Santoro as background survivor couple Nikki Fernandez and Paulo.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, AWARD\n\nText:\nIn addition, three new actors joined the regular cast: Elizabeth Mitchell, as fertility doctor and Other Juliet Burke; and Kiele Sanchez and Rodrigo Santoro as background survivor couple Nikki Fernandez and Paulo.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_177", "text": "On 13 February 2015, scientists (including Geoffrey Marcy, Seth Shostak, Frank Drake and David Brin) at a convention of the American Association for the Advancement of Science, discussed Active SETI and whether transmitting a message to possible intelligent extraterrestrials in the Cosmos was a good idea;", "requested_types": ["SCIENTIST", "ORGANISATION", "PERSON", "MISC"], "distractors": ["PERSON"], "gold": {"SCIENTIST": ["Geoffrey Marcy", "Seth Shostak", "Frank Drake", "David Brin"], "ORGANISATION": ["American Association for the Advancement of Science"], "PERSON": [], "MISC": ["Active SETI", "Cosmos"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ORGANISATION, PERSON, MISC\n\nText:\nOn 13 February 2015, scientists (including Geoffrey Marcy, Seth Shostak, Frank Drake and David Brin) at a convention of the American Association for the Advancement of Science, discussed Active SETI and whether transmitting a message to possible intelligent extraterrestrials in the Cosmos was a good idea;\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ORGANISATION, PERSON, MISC\n\nText:\nOn 13 February 2015, scientists (including Geoffrey Marcy, Seth Shostak, Frank Drake and David Brin) at a convention of the American Association for the Advancement of Science, discussed Active SETI and whether transmitting a message to possible intelligent extraterrestrials in the Cosmos was a good idea;\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_178", "text": "Arizona State University offers undergraduate student housing on four metropolitan Phoenix campuses (Arizona State University Tempe campus, Arizona State University Polytechnic campus, Arizona State University Downtown Phoenix campus, and Arizona State University West campus), plus the ASU Colleges at Lake Havasu City.", "requested_types": ["LOCATION", "UNIVERSITY", "ACADEMIC_JOURNAL"], "distractors": ["ACADEMIC_JOURNAL"], "gold": {"LOCATION": ["Phoenix campuses"], "UNIVERSITY": ["Arizona State University", "Arizona State University Tempe campus", "Arizona State University Polytechnic campus", "Arizona State University Downtown Phoenix campus", "Arizona State University West campus", "ASU Colleges at Lake Havasu City"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, UNIVERSITY, ACADEMIC_JOURNAL\n\nText:\nArizona State University offers undergraduate student housing on four metropolitan Phoenix campuses (Arizona State University Tempe campus, Arizona State University Polytechnic campus, Arizona State University Downtown Phoenix campus, and Arizona State University West campus), plus the ASU Colleges at Lake Havasu City.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, UNIVERSITY, ACADEMIC_JOURNAL\n\nText:\nArizona State University offers undergraduate student housing on four metropolitan Phoenix campuses (Arizona State University Tempe campus, Arizona State University Polytechnic campus, Arizona State University Downtown Phoenix campus, and Arizona State University West campus), plus the ASU Colleges at Lake Havasu City.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_179", "text": "The traditional precursor is N-Nitroso-N-methylurea, but this compound is itself somewhat unstable, and nowadays compounds such as Methylnitronitrosoguanidine (MNNG) and Diazald (Diazald) are preferred.", "requested_types": ["SCIENTIST", "EVENT", "CHEMICAL_COMPOUND", "UNIVERSITY"], "distractors": ["EVENT", "UNIVERSITY", "SCIENTIST"], "gold": {"SCIENTIST": [], "EVENT": [], "CHEMICAL_COMPOUND": ["N-Nitroso-N-methylurea", "Methylnitronitrosoguanidine", "MNNG", "Diazald"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, EVENT, CHEMICAL_COMPOUND, UNIVERSITY\n\nText:\nThe traditional precursor is N-Nitroso-N-methylurea, but this compound is itself somewhat unstable, and nowadays compounds such as Methylnitronitrosoguanidine (MNNG) and Diazald (Diazald) are preferred.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, EVENT, CHEMICAL_COMPOUND, UNIVERSITY\n\nText:\nThe traditional precursor is N-Nitroso-N-methylurea, but this compound is itself somewhat unstable, and nowadays compounds such as Methylnitronitrosoguanidine (MNNG) and Diazald (Diazald) are preferred.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_180", "text": "The International Astronomical Union names all colles (small hills) on Saturn 's moon Titan after characters in Tolkien's work.", "requested_types": ["SCIENTIST", "DISCIPLINE", "ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT", "CHEMICAL_COMPOUND", "ORGANISATION"], "distractors": ["DISCIPLINE", "CHEMICAL_COMPOUND", "CHEMICAL_ELEMENT"], "gold": {"SCIENTIST": ["Tolkien"], "DISCIPLINE": [], "ASTRONOMICAL_OBJECT": ["Saturn", "Titan"], "CHEMICAL_ELEMENT": [], "CHEMICAL_COMPOUND": [], "ORGANISATION": ["International Astronomical Union"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, DISCIPLINE, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, CHEMICAL_COMPOUND, ORGANISATION\n\nText:\nThe International Astronomical Union names all colles (small hills) on Saturn 's moon Titan after characters in Tolkien's work.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, DISCIPLINE, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, CHEMICAL_COMPOUND, ORGANISATION\n\nText:\nThe International Astronomical Union names all colles (small hills) on Saturn 's moon Titan after characters in Tolkien's work.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_181", "text": "Since uranium had been named after the planet Uranus and neptunium after the planet Neptune, element 94 was named after Pluto, which at the time was considered to be a planet as well.", "requested_types": ["ACADEMIC_JOURNAL", "EVENT", "ASTRONOMICAL_OBJECT", "COUNTRY"], "distractors": ["EVENT", "COUNTRY", "ACADEMIC_JOURNAL"], "gold": {"ACADEMIC_JOURNAL": [], "EVENT": [], "ASTRONOMICAL_OBJECT": ["uranium", "Uranus", "neptunium", "Neptune", "Pluto"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, EVENT, ASTRONOMICAL_OBJECT, COUNTRY\n\nText:\nSince uranium had been named after the planet Uranus and neptunium after the planet Neptune, element 94 was named after Pluto, which at the time was considered to be a planet as well.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, EVENT, ASTRONOMICAL_OBJECT, COUNTRY\n\nText:\nSince uranium had been named after the planet Uranus and neptunium after the planet Neptune, element 94 was named after Pluto, which at the time was considered to be a planet as well.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_182", "text": "A pupil of Peter Gustav Lejeune Dirichlet, Gabriel Lamé and Augustin-Louis Cauchy Bjerknes worked for the rest of his life in the field of hydrodynamics.", "requested_types": ["DISCIPLINE", "UNIVERSITY", "THEORY", "SCIENTIST", "ACADEMIC_JOURNAL"], "distractors": ["THEORY", "ACADEMIC_JOURNAL", "UNIVERSITY"], "gold": {"DISCIPLINE": ["hydrodynamics"], "UNIVERSITY": [], "THEORY": [], "SCIENTIST": ["Peter Gustav Lejeune Dirichlet", "Gabriel Lamé", "Augustin-Louis Cauchy Bjerknes"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, UNIVERSITY, THEORY, SCIENTIST, ACADEMIC_JOURNAL\n\nText:\nA pupil of Peter Gustav Lejeune Dirichlet, Gabriel Lamé and Augustin-Louis Cauchy Bjerknes worked for the rest of his life in the field of hydrodynamics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, UNIVERSITY, THEORY, SCIENTIST, ACADEMIC_JOURNAL\n\nText:\nA pupil of Peter Gustav Lejeune Dirichlet, Gabriel Lamé and Augustin-Louis Cauchy Bjerknes worked for the rest of his life in the field of hydrodynamics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_183", "text": "He served as editor of Optics Communications (1987-2006) and was member of the editorial board at Physical Review A.", "requested_types": ["CHEMICAL_COMPOUND", "PERSON", "ACADEMIC_JOURNAL"], "distractors": ["CHEMICAL_COMPOUND", "PERSON"], "gold": {"CHEMICAL_COMPOUND": [], "PERSON": [], "ACADEMIC_JOURNAL": ["Optics Communications", "Physical Review A"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, PERSON, ACADEMIC_JOURNAL\n\nText:\nHe served as editor of Optics Communications (1987-2006) and was member of the editorial board at Physical Review A.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, PERSON, ACADEMIC_JOURNAL\n\nText:\nHe served as editor of Optics Communications (1987-2006) and was member of the editorial board at Physical Review A.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_184", "text": "In the 1840s, Urbain Le Verrier used Newtonian mechanics to predict the position of the then-undiscovered planet Neptune after analyzing perturbations in the orbit of Uranus.", "requested_types": ["DISCIPLINE", "CHEMICAL_COMPOUND", "SCIENTIST", "ORGANISATION", "COUNTRY", "ASTRONOMICAL_OBJECT"], "distractors": ["CHEMICAL_COMPOUND", "ORGANISATION", "COUNTRY"], "gold": {"DISCIPLINE": ["Newtonian mechanics"], "CHEMICAL_COMPOUND": [], "SCIENTIST": ["Urbain Le Verrier"], "ORGANISATION": [], "COUNTRY": [], "ASTRONOMICAL_OBJECT": ["Neptune", "Uranus"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, CHEMICAL_COMPOUND, SCIENTIST, ORGANISATION, COUNTRY, ASTRONOMICAL_OBJECT\n\nText:\nIn the 1840s, Urbain Le Verrier used Newtonian mechanics to predict the position of the then-undiscovered planet Neptune after analyzing perturbations in the orbit of Uranus.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, CHEMICAL_COMPOUND, SCIENTIST, ORGANISATION, COUNTRY, ASTRONOMICAL_OBJECT\n\nText:\nIn the 1840s, Urbain Le Verrier used Newtonian mechanics to predict the position of the then-undiscovered planet Neptune after analyzing perturbations in the orbit of Uranus.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_185", "text": "Esaki moved back to Japan in 1992, subsequently, he served as president of the University of Tsukuba and Shibaura Institute of Technology.", "requested_types": ["ACADEMIC_JOURNAL", "SCIENTIST", "ASTRONOMICAL_OBJECT", "COUNTRY", "UNIVERSITY", "ENZYME"], "distractors": ["ACADEMIC_JOURNAL", "ENZYME", "ASTRONOMICAL_OBJECT"], "gold": {"ACADEMIC_JOURNAL": [], "SCIENTIST": ["Esaki"], "ASTRONOMICAL_OBJECT": [], "COUNTRY": ["Japan"], "UNIVERSITY": ["University of Tsukuba", "Shibaura Institute of Technology"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, ASTRONOMICAL_OBJECT, COUNTRY, UNIVERSITY, ENZYME\n\nText:\nEsaki moved back to Japan in 1992, subsequently, he served as president of the University of Tsukuba and Shibaura Institute of Technology.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, ASTRONOMICAL_OBJECT, COUNTRY, UNIVERSITY, ENZYME\n\nText:\nEsaki moved back to Japan in 1992, subsequently, he served as president of the University of Tsukuba and Shibaura Institute of Technology.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_186", "text": "From 1930 to 1934, Wu studied at National Central University (later renamed Nanjing University), first in mathematics, but later transferring to physics.", "requested_types": ["DISCIPLINE", "ORGANISATION", "UNIVERSITY", "SCIENTIST"], "distractors": ["ORGANISATION"], "gold": {"DISCIPLINE": ["mathematics", "physics"], "ORGANISATION": [], "UNIVERSITY": ["National Central University", "Nanjing University"], "SCIENTIST": ["Wu"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ORGANISATION, UNIVERSITY, SCIENTIST\n\nText:\nFrom 1930 to 1934, Wu studied at National Central University (later renamed Nanjing University), first in mathematics, but later transferring to physics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ORGANISATION, UNIVERSITY, SCIENTIST\n\nText:\nFrom 1930 to 1934, Wu studied at National Central University (later renamed Nanjing University), first in mathematics, but later transferring to physics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_187", "text": "The quarry is at Herniss, to the north of the A 394 road, between Rame and Longdowns.", "requested_types": ["PERSON", "AWARD", "CHEMICAL_ELEMENT", "LOCATION"], "distractors": ["CHEMICAL_ELEMENT", "AWARD", "PERSON"], "gold": {"PERSON": [], "AWARD": [], "CHEMICAL_ELEMENT": [], "LOCATION": ["Herniss", "A 394 road", "Rame", "Longdowns"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, AWARD, CHEMICAL_ELEMENT, LOCATION\n\nText:\nThe quarry is at Herniss, to the north of the A 394 road, between Rame and Longdowns.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, AWARD, CHEMICAL_ELEMENT, LOCATION\n\nText:\nThe quarry is at Herniss, to the north of the A 394 road, between Rame and Longdowns.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_188", "text": "The schemas targeted by emotional selection are those essential for meeting human needs, such as those define by Abraham Maslow and Henry Murray.", "requested_types": ["MISC", "SCIENTIST", "THEORY"], "distractors": ["THEORY"], "gold": {"MISC": ["schemas"], "SCIENTIST": ["Abraham Maslow", "Henry Murray"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST, THEORY\n\nText:\nThe schemas targeted by emotional selection are those essential for meeting human needs, such as those define by Abraham Maslow and Henry Murray.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST, THEORY\n\nText:\nThe schemas targeted by emotional selection are those essential for meeting human needs, such as those define by Abraham Maslow and Henry Murray.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_189", "text": "He was awarded the Louisa Gross Horwitz Prize from Columbia University in 1991, the Louis-Jeantet Prize for Medicine in 1993, the Otto Warburg Medal in 1999 and half of the Nobel Prize in Chemistry in 2002 for his development of nuclear magnetic resonance spectroscopy for determining the three-dimensional structure of biological macromolecules in solution.", "requested_types": ["SCIENTIST", "AWARD", "MISC", "UNIVERSITY"], "distractors": ["MISC", "SCIENTIST"], "gold": {"SCIENTIST": [], "AWARD": ["Louisa Gross Horwitz Prize", "Louis-Jeantet Prize for Medicine", "Otto Warburg Medal", "Nobel Prize in Chemistry", "nuclear magnetic resonance spectroscopy"], "MISC": [], "UNIVERSITY": ["Columbia University"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, MISC, UNIVERSITY\n\nText:\nHe was awarded the Louisa Gross Horwitz Prize from Columbia University in 1991, the Louis-Jeantet Prize for Medicine in 1993, the Otto Warburg Medal in 1999 and half of the Nobel Prize in Chemistry in 2002 for his development of nuclear magnetic resonance spectroscopy for determining the three-dimensional structure of biological macromolecules in solution.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, MISC, UNIVERSITY\n\nText:\nHe was awarded the Louisa Gross Horwitz Prize from Columbia University in 1991, the Louis-Jeantet Prize for Medicine in 1993, the Otto Warburg Medal in 1999 and half of the Nobel Prize in Chemistry in 2002 for his development of nuclear magnetic resonance spectroscopy for determining the three-dimensional structure of biological macromolecules in solution.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_190", "text": "Her group has characterized the role of a number of Pentatricopeptide repeat -containing proteins in gene expression, as well as proteins involved in splicing in the chloroplast, discovering a number of CRM domain -containing proteins and coining the domain name.", "requested_types": ["PROTEIN", "MISC", "ENZYME", "AWARD"], "distractors": ["ENZYME", "AWARD"], "gold": {"PROTEIN": ["Pentatricopeptide repeat -containing proteins", "CRM domain -containing proteins"], "MISC": ["chloroplast"], "ENZYME": [], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, ENZYME, AWARD\n\nText:\nHer group has characterized the role of a number of Pentatricopeptide repeat -containing proteins in gene expression, as well as proteins involved in splicing in the chloroplast, discovering a number of CRM domain -containing proteins and coining the domain name.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, ENZYME, AWARD\n\nText:\nHer group has characterized the role of a number of Pentatricopeptide repeat -containing proteins in gene expression, as well as proteins involved in splicing in the chloroplast, discovering a number of CRM domain -containing proteins and coining the domain name.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_191", "text": "Comparative table of countries with very high human development (same or higher than 0.8), according to UNDP; OECD members; advanced economies, according to the International Monetary Fund; high income economies, according to the World Bank and income per capita (purchasing power parity) higher than $22,000, according to the International Monetary Fund.", "requested_types": ["ORGANISATION", "PROTEIN"], "distractors": ["PROTEIN"], "gold": {"ORGANISATION": ["UNDP", "OECD", "International Monetary Fund", "World Bank"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PROTEIN\n\nText:\nComparative table of countries with very high human development (same or higher than 0.8), according to UNDP; OECD members; advanced economies, according to the International Monetary Fund; high income economies, according to the World Bank and income per capita (purchasing power parity) higher than $22,000, according to the International Monetary Fund.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PROTEIN\n\nText:\nComparative table of countries with very high human development (same or higher than 0.8), according to UNDP; OECD members; advanced economies, according to the International Monetary Fund; high income economies, according to the World Bank and income per capita (purchasing power parity) higher than $22,000, according to the International Monetary Fund.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_192", "text": "Pathogenic spirochetes, including B. burgdorferi and T. pallidum, use their proteolipid Bacterial adhesin s to stick to victim cells.", "requested_types": ["MISC", "THEORY"], "distractors": ["THEORY"], "gold": {"MISC": ["Pathogenic spirochetes", "B. burgdorferi", "T. pallidum", "Bacterial adhesin"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, THEORY\n\nText:\nPathogenic spirochetes, including B. burgdorferi and T. pallidum, use their proteolipid Bacterial adhesin s to stick to victim cells.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, THEORY\n\nText:\nPathogenic spirochetes, including B. burgdorferi and T. pallidum, use their proteolipid Bacterial adhesin s to stick to victim cells.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_193", "text": "The alkaline earth metals are named after their oxide s, the alkaline earths, whose old-fashioned names were beryllia, Magnesium oxide, Calcium oxide, strontia, and baryta.", "requested_types": ["SCIENTIST", "CHEMICAL_COMPOUND", "DISCIPLINE", "UNIVERSITY", "MISC"], "distractors": ["DISCIPLINE", "UNIVERSITY", "SCIENTIST"], "gold": {"SCIENTIST": [], "CHEMICAL_COMPOUND": ["oxide", "beryllia", "Magnesium oxide", "Calcium oxide", "strontia", "baryta"], "DISCIPLINE": [], "UNIVERSITY": [], "MISC": ["alkaline earth metals", "alkaline earths"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_COMPOUND, DISCIPLINE, UNIVERSITY, MISC\n\nText:\nThe alkaline earth metals are named after their oxide s, the alkaline earths, whose old-fashioned names were beryllia, Magnesium oxide, Calcium oxide, strontia, and baryta.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_COMPOUND, DISCIPLINE, UNIVERSITY, MISC\n\nText:\nThe alkaline earth metals are named after their oxide s, the alkaline earths, whose old-fashioned names were beryllia, Magnesium oxide, Calcium oxide, strontia, and baryta.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_194", "text": "The Copenhagen interpretation is an expression of the meaning of quantum mechanics that was largely devised from 1925 to 1927 by Niels Bohr and Werner Heisenberg.", "requested_types": ["PERSON", "LOCATION", "DISCIPLINE", "SCIENTIST"], "distractors": ["PERSON"], "gold": {"PERSON": [], "LOCATION": ["Copenhagen"], "DISCIPLINE": ["quantum mechanics"], "SCIENTIST": ["Niels Bohr", "Werner Heisenberg"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, DISCIPLINE, SCIENTIST\n\nText:\nThe Copenhagen interpretation is an expression of the meaning of quantum mechanics that was largely devised from 1925 to 1927 by Niels Bohr and Werner Heisenberg.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, DISCIPLINE, SCIENTIST\n\nText:\nThe Copenhagen interpretation is an expression of the meaning of quantum mechanics that was largely devised from 1925 to 1927 by Niels Bohr and Werner Heisenberg.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_195", "text": "He studied physics at the University of Heidelberg, the University of Lausanne, and, ultimately, the Humboldt University of Berlin under Max Planck.", "requested_types": ["DISCIPLINE", "ASTRONOMICAL_OBJECT", "UNIVERSITY", "SCIENTIST"], "distractors": ["ASTRONOMICAL_OBJECT"], "gold": {"DISCIPLINE": ["physics"], "ASTRONOMICAL_OBJECT": [], "UNIVERSITY": ["University of Heidelberg", "University of Lausanne", "Humboldt University of Berlin"], "SCIENTIST": ["Max Planck"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ASTRONOMICAL_OBJECT, UNIVERSITY, SCIENTIST\n\nText:\nHe studied physics at the University of Heidelberg, the University of Lausanne, and, ultimately, the Humboldt University of Berlin under Max Planck.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ASTRONOMICAL_OBJECT, UNIVERSITY, SCIENTIST\n\nText:\nHe studied physics at the University of Heidelberg, the University of Lausanne, and, ultimately, the Humboldt University of Berlin under Max Planck.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_196", "text": "He entered the accredited Illinois Central College, then transferred to the unaccredited Midwestern Baptist College in 1972, attaining a Bachelor of Religious Education in 1974.", "requested_types": ["DISCIPLINE", "ASTRONOMICAL_OBJECT", "UNIVERSITY", "MISC"], "distractors": ["ASTRONOMICAL_OBJECT"], "gold": {"DISCIPLINE": ["Religious Education"], "ASTRONOMICAL_OBJECT": [], "UNIVERSITY": ["Illinois Central College", "Midwestern Baptist College"], "MISC": ["Bachelor"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ASTRONOMICAL_OBJECT, UNIVERSITY, MISC\n\nText:\nHe entered the accredited Illinois Central College, then transferred to the unaccredited Midwestern Baptist College in 1972, attaining a Bachelor of Religious Education in 1974.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ASTRONOMICAL_OBJECT, UNIVERSITY, MISC\n\nText:\nHe entered the accredited Illinois Central College, then transferred to the unaccredited Midwestern Baptist College in 1972, attaining a Bachelor of Religious Education in 1974.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_197", "text": "The Chun-Tsung scholarships, supervised by the United Board for Christian Higher Education in Asia (New York), are awarded to undergraduates, usually in their 2nd or 3rd year, at six universities, which are Shanghai Jiaotong University, Fudan University, Lanzhou University, Soochow University, Peking University and Taiwan National Tsing Hua University.", "requested_types": ["ORGANISATION", "PERSON", "UNIVERSITY", "THEORY", "AWARD", "LOCATION", "ACADEMIC_JOURNAL"], "distractors": ["PERSON", "THEORY", "ACADEMIC_JOURNAL"], "gold": {"ORGANISATION": ["United Board for Christian Higher Education in Asia"], "PERSON": [], "UNIVERSITY": ["Shanghai Jiaotong University", "Fudan University", "Lanzhou University", "Soochow University", "Peking University", "Taiwan National Tsing Hua University"], "THEORY": [], "AWARD": ["Chun-Tsung scholarships"], "LOCATION": ["New York"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PERSON, UNIVERSITY, THEORY, AWARD, LOCATION, ACADEMIC_JOURNAL\n\nText:\nThe Chun-Tsung scholarships, supervised by the United Board for Christian Higher Education in Asia (New York), are awarded to undergraduates, usually in their 2nd or 3rd year, at six universities, which are Shanghai Jiaotong University, Fudan University, Lanzhou University, Soochow University, Peking University and Taiwan National Tsing Hua University.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PERSON, UNIVERSITY, THEORY, AWARD, LOCATION, ACADEMIC_JOURNAL\n\nText:\nThe Chun-Tsung scholarships, supervised by the United Board for Christian Higher Education in Asia (New York), are awarded to undergraduates, usually in their 2nd or 3rd year, at six universities, which are Shanghai Jiaotong University, Fudan University, Lanzhou University, Soochow University, Peking University and Taiwan National Tsing Hua University.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_198", "text": "This leads to enhanced microbicidal capacity and secretion of high levels of pro-inflammatory cytokines: e.g. IFN-γ, IL-1, IL-6, Interleukin 12, Interleukin 23 and TNFα.", "requested_types": ["CHEMICAL_ELEMENT", "MISC", "ORGANISATION"], "distractors": ["ORGANISATION", "CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_ELEMENT": [], "MISC": ["IFN-γ", "IL-1", "IL-6", "Interleukin 12", "Interleukin 23", "TNFα"], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, MISC, ORGANISATION\n\nText:\nThis leads to enhanced microbicidal capacity and secretion of high levels of pro-inflammatory cytokines: e.g. IFN-γ, IL-1, IL-6, Interleukin 12, Interleukin 23 and TNFα.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, MISC, ORGANISATION\n\nText:\nThis leads to enhanced microbicidal capacity and secretion of high levels of pro-inflammatory cytokines: e.g. IFN-γ, IL-1, IL-6, Interleukin 12, Interleukin 23 and TNFα.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_199", "text": "He was appointed as assistant to the astronomer Giovanni Battista Donati in 1862, professor at the technological institute of Bologna in 1864, professor of physics at the University of Cagliari in 1873, and, finally, successor to his father in 1881 in the chair of technological physics at the University of Pisa.", "requested_types": ["PERSON", "LOCATION", "SCIENTIST", "DISCIPLINE", "UNIVERSITY", "EVENT"], "distractors": ["EVENT", "PERSON"], "gold": {"PERSON": [], "LOCATION": ["technological institute of Bologna"], "SCIENTIST": ["Giovanni Battista Donati"], "DISCIPLINE": ["physics", "technological physics"], "UNIVERSITY": ["University of Cagliari", "University of Pisa"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, SCIENTIST, DISCIPLINE, UNIVERSITY, EVENT\n\nText:\nHe was appointed as assistant to the astronomer Giovanni Battista Donati in 1862, professor at the technological institute of Bologna in 1864, professor of physics at the University of Cagliari in 1873, and, finally, successor to his father in 1881 in the chair of technological physics at the University of Pisa.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, SCIENTIST, DISCIPLINE, UNIVERSITY, EVENT\n\nText:\nHe was appointed as assistant to the astronomer Giovanni Battista Donati in 1862, professor at the technological institute of Bologna in 1864, professor of physics at the University of Cagliari in 1873, and, finally, successor to his father in 1881 in the chair of technological physics at the University of Pisa.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_200", "text": "The Arts Centre Melbourne (which includes the State Theatre, Hamer Hall, the Playhouse and the Fairfax Studio), and the Melbourne Recital Centre are located just to the south of the CBD, with the Sidney Myer Music Bowl in parklands to the east.", "requested_types": ["UNIVERSITY", "THEORY", "LOCATION", "ORGANISATION", "PROTEIN"], "distractors": ["UNIVERSITY", "PROTEIN", "THEORY"], "gold": {"UNIVERSITY": [], "THEORY": [], "LOCATION": ["State Theatre", "Hamer Hall", "Playhouse", "Fairfax Studio", "Melbourne Recital Centre", "CBD", "Sidney Myer Music Bowl", "parklands"], "ORGANISATION": ["Arts Centre Melbourne"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, THEORY, LOCATION, ORGANISATION, PROTEIN\n\nText:\nThe Arts Centre Melbourne (which includes the State Theatre, Hamer Hall, the Playhouse and the Fairfax Studio), and the Melbourne Recital Centre are located just to the south of the CBD, with the Sidney Myer Music Bowl in parklands to the east.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, THEORY, LOCATION, ORGANISATION, PROTEIN\n\nText:\nThe Arts Centre Melbourne (which includes the State Theatre, Hamer Hall, the Playhouse and the Fairfax Studio), and the Melbourne Recital Centre are located just to the south of the CBD, with the Sidney Myer Music Bowl in parklands to the east.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_201", "text": "In addition, he is the founding editor of Theoretical Population Biology (1971-2013) and an associate editor of Genetics, Human Genetics, Annals of Human Genetics, Annals of Human Biology, and Complexity.", "requested_types": ["COUNTRY", "THEORY", "ACADEMIC_JOURNAL"], "distractors": ["THEORY", "COUNTRY"], "gold": {"COUNTRY": [], "THEORY": [], "ACADEMIC_JOURNAL": ["Theoretical Population Biology", "Genetics", "Human Genetics", "Annals of Human Genetics", "Annals of Human Biology", "Complexity"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, THEORY, ACADEMIC_JOURNAL\n\nText:\nIn addition, he is the founding editor of Theoretical Population Biology (1971-2013) and an associate editor of Genetics, Human Genetics, Annals of Human Genetics, Annals of Human Biology, and Complexity.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, THEORY, ACADEMIC_JOURNAL\n\nText:\nIn addition, he is the founding editor of Theoretical Population Biology (1971-2013) and an associate editor of Genetics, Human Genetics, Annals of Human Genetics, Annals of Human Biology, and Complexity.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_202", "text": "Marsden was discovered on 24 March 1971, by Dutch astronomer couple Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plates taken by Dutch-American astronomer Tom Gehrels at Palomar Observatory, California.", "requested_types": ["PERSON", "SCIENTIST", "LOCATION", "ORGANISATION", "ASTRONOMICAL_OBJECT", "MISC"], "distractors": ["PERSON", "ORGANISATION"], "gold": {"PERSON": [], "SCIENTIST": ["Ingrid van Houten-Groeneveld", "Cornelis van Houten", "Tom Gehrels"], "LOCATION": ["Leiden", "Palomar Observatory", "California"], "ORGANISATION": [], "ASTRONOMICAL_OBJECT": ["Marsden"], "MISC": ["Dutch"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, SCIENTIST, LOCATION, ORGANISATION, ASTRONOMICAL_OBJECT, MISC\n\nText:\nMarsden was discovered on 24 March 1971, by Dutch astronomer couple Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plates taken by Dutch-American astronomer Tom Gehrels at Palomar Observatory, California.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, SCIENTIST, LOCATION, ORGANISATION, ASTRONOMICAL_OBJECT, MISC\n\nText:\nMarsden was discovered on 24 March 1971, by Dutch astronomer couple Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plates taken by Dutch-American astronomer Tom Gehrels at Palomar Observatory, California.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_203", "text": "Jeffrey S. Moore has published over 300 articles in journals such as Macromolecules, the Journal of Chemical Education, Advanced Materials and the Journal of Materials Chemistry.", "requested_types": ["SCIENTIST", "AWARD", "ACADEMIC_JOURNAL", "LOCATION"], "distractors": ["AWARD", "LOCATION"], "gold": {"SCIENTIST": ["Jeffrey S. Moore"], "AWARD": [], "ACADEMIC_JOURNAL": ["Macromolecules", "Journal of Chemical Education", "Advanced Materials", "Journal of Materials Chemistry"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, ACADEMIC_JOURNAL, LOCATION\n\nText:\nJeffrey S. Moore has published over 300 articles in journals such as Macromolecules, the Journal of Chemical Education, Advanced Materials and the Journal of Materials Chemistry.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, ACADEMIC_JOURNAL, LOCATION\n\nText:\nJeffrey S. Moore has published over 300 articles in journals such as Macromolecules, the Journal of Chemical Education, Advanced Materials and the Journal of Materials Chemistry.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_204", "text": "In addition, scientific results obtained using HAARP are routinely published in major research journals (such as Geophysical Research Letters and Journal of Geophysical Research), written both by university scientists (American and foreign) and by U.S. Department of Defense research lab scientists.", "requested_types": ["ORGANISATION", "ACADEMIC_JOURNAL", "MISC", "AWARD"], "distractors": ["AWARD"], "gold": {"ORGANISATION": ["U.S. Department of Defense research lab"], "ACADEMIC_JOURNAL": ["Geophysical Research Letters", "Journal of Geophysical Research"], "MISC": ["HAARP", "American"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ACADEMIC_JOURNAL, MISC, AWARD\n\nText:\nIn addition, scientific results obtained using HAARP are routinely published in major research journals (such as Geophysical Research Letters and Journal of Geophysical Research), written both by university scientists (American and foreign) and by U.S. Department of Defense research lab scientists.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ACADEMIC_JOURNAL, MISC, AWARD\n\nText:\nIn addition, scientific results obtained using HAARP are routinely published in major research journals (such as Geophysical Research Letters and Journal of Geophysical Research), written both by university scientists (American and foreign) and by U.S. Department of Defense research lab scientists.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_205", "text": "His work has appeared in leading scientific journals such as Nature, Science, Nature Genetics, Molecular Biology and Evolution, Journal of Evolutionary Biology, as well as popular magazines such as Scientific American.", "requested_types": ["DISCIPLINE", "ACADEMIC_JOURNAL"], "distractors": ["DISCIPLINE"], "gold": {"DISCIPLINE": [], "ACADEMIC_JOURNAL": ["Nature", "Science", "Nature Genetics", "Molecular Biology and Evolution", "Journal of Evolutionary Biology", "Scientific American"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ACADEMIC_JOURNAL\n\nText:\nHis work has appeared in leading scientific journals such as Nature, Science, Nature Genetics, Molecular Biology and Evolution, Journal of Evolutionary Biology, as well as popular magazines such as Scientific American.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ACADEMIC_JOURNAL\n\nText:\nHis work has appeared in leading scientific journals such as Nature, Science, Nature Genetics, Molecular Biology and Evolution, Journal of Evolutionary Biology, as well as popular magazines such as Scientific American.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_206", "text": "The city is home to the Department of Physical Education and Sport Science of the Aristotle University of Thessaloniki () and the Serres Campus of the International Hellenic University (former Technological Educational Institute of Central Macedonia), composed of the Faculty of Engineering, the Faculty of Economics and Management, and the Department of Interior Architecture and Design.", "requested_types": ["UNIVERSITY", "EVENT", "DISCIPLINE", "ORGANISATION", "COUNTRY"], "distractors": ["COUNTRY", "EVENT", "DISCIPLINE"], "gold": {"UNIVERSITY": ["Aristotle University of Thessaloniki", "Serres Campus of the International Hellenic University", "Technological Educational Institute of Central Macedonia"], "EVENT": [], "DISCIPLINE": [], "ORGANISATION": ["Department of Physical Education and Sport Science", "Faculty of Engineering", "Faculty of Economics and Management", "Department of Interior Architecture and Design"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, EVENT, DISCIPLINE, ORGANISATION, COUNTRY\n\nText:\nThe city is home to the Department of Physical Education and Sport Science of the Aristotle University of Thessaloniki () and the Serres Campus of the International Hellenic University (former Technological Educational Institute of Central Macedonia), composed of the Faculty of Engineering, the Faculty of Economics and Management, and the Department of Interior Architecture and Design.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, EVENT, DISCIPLINE, ORGANISATION, COUNTRY\n\nText:\nThe city is home to the Department of Physical Education and Sport Science of the Aristotle University of Thessaloniki () and the Serres Campus of the International Hellenic University (former Technological Educational Institute of Central Macedonia), composed of the Faculty of Engineering, the Faculty of Economics and Management, and the Department of Interior Architecture and Design.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_207", "text": "Bangladesh is a member of the UN, World Trade Organization, International Monetary Fund, the World Bank, Asian Development Bank, OIC, Islamic Development Bank, SAARC, BIMSTEC and the Islamic Military Counter Terrorism Coalition.", "requested_types": ["DISCIPLINE", "ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT", "COUNTRY", "ORGANISATION"], "distractors": ["DISCIPLINE", "CHEMICAL_ELEMENT", "ASTRONOMICAL_OBJECT"], "gold": {"DISCIPLINE": [], "ASTRONOMICAL_OBJECT": [], "CHEMICAL_ELEMENT": [], "COUNTRY": ["Bangladesh"], "ORGANISATION": ["UN", "World Trade Organization", "International Monetary Fund", "World Bank", "Asian Development Bank", "OIC", "Islamic Development Bank", "SAARC", "BIMSTEC", "Islamic Military Counter Terrorism Coalition"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, COUNTRY, ORGANISATION\n\nText:\nBangladesh is a member of the UN, World Trade Organization, International Monetary Fund, the World Bank, Asian Development Bank, OIC, Islamic Development Bank, SAARC, BIMSTEC and the Islamic Military Counter Terrorism Coalition.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, COUNTRY, ORGANISATION\n\nText:\nBangladesh is a member of the UN, World Trade Organization, International Monetary Fund, the World Bank, Asian Development Bank, OIC, Islamic Development Bank, SAARC, BIMSTEC and the Islamic Military Counter Terrorism Coalition.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_208", "text": "Flügge also extended Niels Bohr and John Archibald Wheeler theory of nuclear fission published in 1939. Niels Bohr and J. A. Wheeler Mechanism of nuclear fission, Phys.", "requested_types": ["PERSON", "DISCIPLINE", "THEORY", "ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT", "SCIENTIST", "MISC"], "distractors": ["CHEMICAL_ELEMENT", "PERSON", "ASTRONOMICAL_OBJECT"], "gold": {"PERSON": [], "DISCIPLINE": ["Phys"], "THEORY": ["theory of nuclear fission"], "ASTRONOMICAL_OBJECT": [], "CHEMICAL_ELEMENT": [], "SCIENTIST": ["Flügge", "Niels Bohr", "John Archibald Wheeler", "J. A. Wheeler"], "MISC": ["nuclear fission"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, DISCIPLINE, THEORY, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, SCIENTIST, MISC\n\nText:\nFlügge also extended Niels Bohr and John Archibald Wheeler theory of nuclear fission published in 1939. Niels Bohr and J. A. Wheeler Mechanism of nuclear fission, Phys.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, DISCIPLINE, THEORY, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, SCIENTIST, MISC\n\nText:\nFlügge also extended Niels Bohr and John Archibald Wheeler theory of nuclear fission published in 1939. Niels Bohr and J. A. Wheeler Mechanism of nuclear fission, Phys.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_209", "text": "His work won him the Rumford Medal of the Royal Society in 1838, and in 1843 he received its Royal Medal for a paper on the Transparency of the Atmosphere and the Laws of Extinction of the Sun's Rays passing through it.", "requested_types": ["DISCIPLINE", "SCIENTIST", "ASTRONOMICAL_OBJECT", "AWARD"], "distractors": ["DISCIPLINE", "SCIENTIST"], "gold": {"DISCIPLINE": [], "SCIENTIST": [], "ASTRONOMICAL_OBJECT": ["Sun"], "AWARD": ["Rumford Medal of the Royal Society", "Royal Medal"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, SCIENTIST, ASTRONOMICAL_OBJECT, AWARD\n\nText:\nHis work won him the Rumford Medal of the Royal Society in 1838, and in 1843 he received its Royal Medal for a paper on the Transparency of the Atmosphere and the Laws of Extinction of the Sun's Rays passing through it.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, SCIENTIST, ASTRONOMICAL_OBJECT, AWARD\n\nText:\nHis work won him the Rumford Medal of the Royal Society in 1838, and in 1843 he received its Royal Medal for a paper on the Transparency of the Atmosphere and the Laws of Extinction of the Sun's Rays passing through it.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_210", "text": "The Msub2 / sub muscarinic receptors decrease the heart rate by inhibiting depolarization of the sinoatrial node via Gi alpha subunit - G protein-coupled receptor and through modulation of muscarinic potassium channels.", "requested_types": ["PROTEIN", "MISC", "CHEMICAL_COMPOUND", "DISCIPLINE"], "distractors": ["DISCIPLINE"], "gold": {"PROTEIN": ["Gi alpha subunit", "G protein-coupled receptor"], "MISC": ["depolarization", "sinoatrial node", "muscarinic potassium channels"], "CHEMICAL_COMPOUND": ["Msub2 / sub muscarinic receptors"], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, CHEMICAL_COMPOUND, DISCIPLINE\n\nText:\nThe Msub2 / sub muscarinic receptors decrease the heart rate by inhibiting depolarization of the sinoatrial node via Gi alpha subunit - G protein-coupled receptor and through modulation of muscarinic potassium channels.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, CHEMICAL_COMPOUND, DISCIPLINE\n\nText:\nThe Msub2 / sub muscarinic receptors decrease the heart rate by inhibiting depolarization of the sinoatrial node via Gi alpha subunit - G protein-coupled receptor and through modulation of muscarinic potassium channels.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_211", "text": "Employing a starting feedstock mixture of 1-Butene and 2-Butene, which is known as Raffinate-2, produces phenol and a mixture of acetone and butanone instead of only phenol and acetone in the original.", "requested_types": ["ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND", "MISC"], "distractors": ["ASTRONOMICAL_OBJECT"], "gold": {"ASTRONOMICAL_OBJECT": [], "CHEMICAL_COMPOUND": ["1-Butene", "2-Butene", "phenol", "acetone", "butanone"], "MISC": ["Raffinate-2"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, MISC\n\nText:\nEmploying a starting feedstock mixture of 1-Butene and 2-Butene, which is known as Raffinate-2, produces phenol and a mixture of acetone and butanone instead of only phenol and acetone in the original.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, MISC\n\nText:\nEmploying a starting feedstock mixture of 1-Butene and 2-Butene, which is known as Raffinate-2, produces phenol and a mixture of acetone and butanone instead of only phenol and acetone in the original.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_212", "text": "Kostiuk and his students were first to measure calcium currents through neuron soma, discover ATP-receptors and Acid-sensing ion channel currents.", "requested_types": ["MISC", "THEORY", "CHEMICAL_ELEMENT", "EVENT", "SCIENTIST"], "distractors": ["THEORY", "EVENT"], "gold": {"MISC": ["neuron soma", "ATP-receptors", "Acid-sensing ion channel"], "THEORY": [], "CHEMICAL_ELEMENT": ["calcium"], "EVENT": [], "SCIENTIST": ["Kostiuk"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, THEORY, CHEMICAL_ELEMENT, EVENT, SCIENTIST\n\nText:\nKostiuk and his students were first to measure calcium currents through neuron soma, discover ATP-receptors and Acid-sensing ion channel currents.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, THEORY, CHEMICAL_ELEMENT, EVENT, SCIENTIST\n\nText:\nKostiuk and his students were first to measure calcium currents through neuron soma, discover ATP-receptors and Acid-sensing ion channel currents.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_213", "text": "In 1944, Folch was appointed director of the new Biological Research Laboratory at the McLean Hospital (a division of Massachusetts General Hospital) and assistant professor of biological chemistry at Harvard Medical School to develop a program in Neuroscience.", "requested_types": ["SCIENTIST", "DISCIPLINE", "ORGANISATION", "CHEMICAL_COMPOUND", "LOCATION", "CHEMICAL_ELEMENT", "ACADEMIC_JOURNAL", "UNIVERSITY"], "distractors": ["CHEMICAL_COMPOUND", "LOCATION", "CHEMICAL_ELEMENT"], "gold": {"SCIENTIST": ["Folch"], "DISCIPLINE": ["biological chemistry"], "ORGANISATION": ["Biological Research Laboratory", "McLean Hospital", "Massachusetts General Hospital"], "CHEMICAL_COMPOUND": [], "LOCATION": [], "CHEMICAL_ELEMENT": [], "ACADEMIC_JOURNAL": ["Neuroscience"], "UNIVERSITY": ["Harvard Medical School"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, DISCIPLINE, ORGANISATION, CHEMICAL_COMPOUND, LOCATION, CHEMICAL_ELEMENT, ACADEMIC_JOURNAL, UNIVERSITY\n\nText:\nIn 1944, Folch was appointed director of the new Biological Research Laboratory at the McLean Hospital (a division of Massachusetts General Hospital) and assistant professor of biological chemistry at Harvard Medical School to develop a program in Neuroscience.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, DISCIPLINE, ORGANISATION, CHEMICAL_COMPOUND, LOCATION, CHEMICAL_ELEMENT, ACADEMIC_JOURNAL, UNIVERSITY\n\nText:\nIn 1944, Folch was appointed director of the new Biological Research Laboratory at the McLean Hospital (a division of Massachusetts General Hospital) and assistant professor of biological chemistry at Harvard Medical School to develop a program in Neuroscience.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_214", "text": "The Right Stuff won four Academy Awards: for Academy Award for Best Sound Editing (Jay Boekelheide), for Academy Award for Best Film Editing, for Academy Award for Best Original Score and for Academy Award for Best Sound Mixing (Mark Berger, Tom Scott, Randy Thom and David MacMillan).", "requested_types": ["AWARD", "MISC", "LOCATION", "THEORY", "PERSON", "ASTRONOMICAL_OBJECT"], "distractors": ["LOCATION", "THEORY", "ASTRONOMICAL_OBJECT"], "gold": {"AWARD": ["Academy Award for Best Sound Editing", "Academy Award for Best Film Editing", "Academy Award for Best Original Score", "Academy Award for Best Sound Mixing"], "MISC": ["The Right Stuff"], "LOCATION": [], "THEORY": [], "PERSON": ["Jay Boekelheide", "Mark Berger", "Tom Scott", "Randy Thom", "David MacMillan"], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, MISC, LOCATION, THEORY, PERSON, ASTRONOMICAL_OBJECT\n\nText:\nThe Right Stuff won four Academy Awards: for Academy Award for Best Sound Editing (Jay Boekelheide), for Academy Award for Best Film Editing, for Academy Award for Best Original Score and for Academy Award for Best Sound Mixing (Mark Berger, Tom Scott, Randy Thom and David MacMillan).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, MISC, LOCATION, THEORY, PERSON, ASTRONOMICAL_OBJECT\n\nText:\nThe Right Stuff won four Academy Awards: for Academy Award for Best Sound Editing (Jay Boekelheide), for Academy Award for Best Film Editing, for Academy Award for Best Original Score and for Academy Award for Best Sound Mixing (Mark Berger, Tom Scott, Randy Thom and David MacMillan).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_215", "text": "It was discovered on 16 October 1977, by Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, and Tom Gehrels at Palomar Observatory in California.", "requested_types": ["SCIENTIST", "LOCATION", "ENZYME", "MISC"], "distractors": ["MISC", "ENZYME"], "gold": {"SCIENTIST": ["Ingrid van Houten-Groeneveld", "Cornelis van Houten", "Tom Gehrels"], "LOCATION": ["Leiden", "Palomar Observatory", "California"], "ENZYME": [], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, LOCATION, ENZYME, MISC\n\nText:\nIt was discovered on 16 October 1977, by Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, and Tom Gehrels at Palomar Observatory in California.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, LOCATION, ENZYME, MISC\n\nText:\nIt was discovered on 16 October 1977, by Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, and Tom Gehrels at Palomar Observatory in California.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_216", "text": "The Royal Society elected him a Fellow of the Royal Society in June 1877 and bestowed their Rumford Medal (1894), Davy (1909), and Copley Medal (1916) medals upon him for his work, as well as inviting him to deliver their Bakerian Lecture in 1901.", "requested_types": ["AWARD", "ORGANISATION", "ASTRONOMICAL_OBJECT", "LOCATION", "CHEMICAL_ELEMENT", "MISC"], "distractors": ["CHEMICAL_ELEMENT", "ASTRONOMICAL_OBJECT", "LOCATION"], "gold": {"AWARD": ["Fellow of the Royal Society", "Rumford Medal", "Davy", "Copley Medal"], "ORGANISATION": ["Royal Society"], "ASTRONOMICAL_OBJECT": [], "LOCATION": [], "CHEMICAL_ELEMENT": [], "MISC": ["Bakerian Lecture"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ORGANISATION, ASTRONOMICAL_OBJECT, LOCATION, CHEMICAL_ELEMENT, MISC\n\nText:\nThe Royal Society elected him a Fellow of the Royal Society in June 1877 and bestowed their Rumford Medal (1894), Davy (1909), and Copley Medal (1916) medals upon him for his work, as well as inviting him to deliver their Bakerian Lecture in 1901.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ORGANISATION, ASTRONOMICAL_OBJECT, LOCATION, CHEMICAL_ELEMENT, MISC\n\nText:\nThe Royal Society elected him a Fellow of the Royal Society in June 1877 and bestowed their Rumford Medal (1894), Davy (1909), and Copley Medal (1916) medals upon him for his work, as well as inviting him to deliver their Bakerian Lecture in 1901.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_217", "text": "Witten has been honored with numerous awards including a MacArthur Grant (1982), the Fields Medal (1990), the Nemmers Prize in Mathematics (2000), the National Medal of Science, retrieved 2013-09-01.", "requested_types": ["UNIVERSITY", "AWARD", "SCIENTIST", "ENZYME", "ORGANISATION"], "distractors": ["UNIVERSITY", "ORGANISATION", "ENZYME"], "gold": {"UNIVERSITY": [], "AWARD": ["MacArthur Grant", "Fields Medal", "Nemmers Prize in Mathematics", "National Medal of Science"], "SCIENTIST": ["Witten"], "ENZYME": [], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, AWARD, SCIENTIST, ENZYME, ORGANISATION\n\nText:\nWitten has been honored with numerous awards including a MacArthur Grant (1982), the Fields Medal (1990), the Nemmers Prize in Mathematics (2000), the National Medal of Science, retrieved 2013-09-01.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, AWARD, SCIENTIST, ENZYME, ORGANISATION\n\nText:\nWitten has been honored with numerous awards including a MacArthur Grant (1982), the Fields Medal (1990), the Nemmers Prize in Mathematics (2000), the National Medal of Science, retrieved 2013-09-01.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_218", "text": "2060 Chiron is a small Solar System body in the outer Solar System, orbit ing the Sun between Saturn and Uranus.", "requested_types": ["ASTRONOMICAL_OBJECT", "MISC", "UNIVERSITY"], "distractors": ["UNIVERSITY"], "gold": {"ASTRONOMICAL_OBJECT": ["2060 Chiron", "Sun", "Saturn", "Uranus"], "MISC": ["small Solar System body", "Solar System"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, UNIVERSITY\n\nText:\n2060 Chiron is a small Solar System body in the outer Solar System, orbit ing the Sun between Saturn and Uranus.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, UNIVERSITY\n\nText:\n2060 Chiron is a small Solar System body in the outer Solar System, orbit ing the Sun between Saturn and Uranus.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_219", "text": "Robert Hooke, Mikhail Lomonosov, Ole Borch, and Pierre Bayden all successfully created oxygen, but did not realize it at the time.", "requested_types": ["ORGANISATION", "ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND", "SCIENTIST"], "distractors": ["ORGANISATION", "ASTRONOMICAL_OBJECT"], "gold": {"ORGANISATION": [], "ASTRONOMICAL_OBJECT": [], "CHEMICAL_COMPOUND": ["oxygen"], "SCIENTIST": ["Robert Hooke", "Mikhail Lomonosov", "Ole Borch", "Pierre Bayden"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, SCIENTIST\n\nText:\nRobert Hooke, Mikhail Lomonosov, Ole Borch, and Pierre Bayden all successfully created oxygen, but did not realize it at the time.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, SCIENTIST\n\nText:\nRobert Hooke, Mikhail Lomonosov, Ole Borch, and Pierre Bayden all successfully created oxygen, but did not realize it at the time.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_220", "text": "In 1753, he was elected a foreign member of the Royal Swedish Academy of Sciences, and on 15 December 1769 a foreign member of the Royal Danish Academy of Sciences and Letters.", "requested_types": ["THEORY", "PERSON", "PROTEIN", "ORGANISATION"], "distractors": ["THEORY", "PERSON", "PROTEIN"], "gold": {"THEORY": [], "PERSON": [], "PROTEIN": [], "ORGANISATION": ["Royal Swedish Academy of Sciences", "Royal Danish Academy of Sciences and Letters"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, PERSON, PROTEIN, ORGANISATION\n\nText:\nIn 1753, he was elected a foreign member of the Royal Swedish Academy of Sciences, and on 15 December 1769 a foreign member of the Royal Danish Academy of Sciences and Letters.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, PERSON, PROTEIN, ORGANISATION\n\nText:\nIn 1753, he was elected a foreign member of the Royal Swedish Academy of Sciences, and on 15 December 1769 a foreign member of the Royal Danish Academy of Sciences and Letters.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_221", "text": "A research into the mutual perturbations of Jupiter and Saturn secured for him the prize of the Berlin Academy in 1830, and a memoir on cometary disturbances was crowned by the Paris Academy in 1850.", "requested_types": ["ASTRONOMICAL_OBJECT", "ORGANISATION", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"ASTRONOMICAL_OBJECT": ["Jupiter", "Saturn"], "ORGANISATION": ["Berlin Academy", "Paris Academy"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ORGANISATION, COUNTRY\n\nText:\nA research into the mutual perturbations of Jupiter and Saturn secured for him the prize of the Berlin Academy in 1830, and a memoir on cometary disturbances was crowned by the Paris Academy in 1850.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ORGANISATION, COUNTRY\n\nText:\nA research into the mutual perturbations of Jupiter and Saturn secured for him the prize of the Berlin Academy in 1830, and a memoir on cometary disturbances was crowned by the Paris Academy in 1850.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_222", "text": "The lithophile elements include: Al, Boron, Barium, Beryllium, Bromine, Calcium, Chlorine, Cr, Caesium, Fluorine, Iodine, Hf, Potassium, Lithium, Magnesium, Sodium, Niobium, O, P, Rubidium, Sc, Si, Strontium, Ta, Th, Ti, U, V, Y, Zirconium, W and the lanthanides.", "requested_types": ["CHEMICAL_COMPOUND", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_COMPOUND"], "gold": {"CHEMICAL_COMPOUND": [], "CHEMICAL_ELEMENT": ["Al", "Boron", "Barium", "Beryllium", "Bromine", "Calcium", "Chlorine", "Cr", "Caesium", "Fluorine", "Iodine", "Hf", "Potassium", "Lithium", "Magnesium", "Sodium", "Niobium", "O", "P", "Rubidium", "Sc", "Si", "Strontium", "Ta", "Th", "Ti", "U", "V", "Y", "Zirconium", "W", "lanthanides"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, CHEMICAL_ELEMENT\n\nText:\nThe lithophile elements include: Al, Boron, Barium, Beryllium, Bromine, Calcium, Chlorine, Cr, Caesium, Fluorine, Iodine, Hf, Potassium, Lithium, Magnesium, Sodium, Niobium, O, P, Rubidium, Sc, Si, Strontium, Ta, Th, Ti, U, V, Y, Zirconium, W and the lanthanides.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, CHEMICAL_ELEMENT\n\nText:\nThe lithophile elements include: Al, Boron, Barium, Beryllium, Bromine, Calcium, Chlorine, Cr, Caesium, Fluorine, Iodine, Hf, Potassium, Lithium, Magnesium, Sodium, Niobium, O, P, Rubidium, Sc, Si, Strontium, Ta, Th, Ti, U, V, Y, Zirconium, W and the lanthanides.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_223", "text": "He has been elected to the US National Academy of Sciences and Britain's Royal Society, as well as to the American Philosophical Society and the American Academy of Arts and Sciences.", "requested_types": ["ACADEMIC_JOURNAL", "SCIENTIST", "AWARD", "LOCATION", "ORGANISATION"], "distractors": ["AWARD", "ACADEMIC_JOURNAL", "SCIENTIST"], "gold": {"ACADEMIC_JOURNAL": [], "SCIENTIST": [], "AWARD": [], "LOCATION": ["Britain"], "ORGANISATION": ["US National Academy of Sciences", "Royal Society", "American Philosophical Society", "American Academy of Arts and Sciences"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, AWARD, LOCATION, ORGANISATION\n\nText:\nHe has been elected to the US National Academy of Sciences and Britain's Royal Society, as well as to the American Philosophical Society and the American Academy of Arts and Sciences.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, AWARD, LOCATION, ORGANISATION\n\nText:\nHe has been elected to the US National Academy of Sciences and Britain's Royal Society, as well as to the American Philosophical Society and the American Academy of Arts and Sciences.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_224", "text": "After receiving his D. Sc. in physics from Humboldt University of Berlin in Berlin, Germany in 1938, Peterlin accepted in 1939 the chair as a professor of physics at the University of Ljubljana, where he remained for 22 years.", "requested_types": ["DISCIPLINE", "COUNTRY", "UNIVERSITY", "PERSON", "LOCATION", "PROTEIN", "EVENT"], "distractors": ["EVENT", "PROTEIN"], "gold": {"DISCIPLINE": ["physics"], "COUNTRY": ["Germany"], "UNIVERSITY": ["Humboldt University of Berlin", "University of Ljubljana"], "PERSON": ["Peterlin"], "LOCATION": ["Berlin"], "PROTEIN": [], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, COUNTRY, UNIVERSITY, PERSON, LOCATION, PROTEIN, EVENT\n\nText:\nAfter receiving his D. Sc. in physics from Humboldt University of Berlin in Berlin, Germany in 1938, Peterlin accepted in 1939 the chair as a professor of physics at the University of Ljubljana, where he remained for 22 years.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, COUNTRY, UNIVERSITY, PERSON, LOCATION, PROTEIN, EVENT\n\nText:\nAfter receiving his D. Sc. in physics from Humboldt University of Berlin in Berlin, Germany in 1938, Peterlin accepted in 1939 the chair as a professor of physics at the University of Ljubljana, where he remained for 22 years.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_225", "text": "Histone s, gene regulatory proteins, DNA polymerase and RNA polymerase s, and other substances essential for nuclear activities must be imported from the cytoplasm.", "requested_types": ["PROTEIN", "ASTRONOMICAL_OBJECT", "ENZYME", "MISC"], "distractors": ["ASTRONOMICAL_OBJECT"], "gold": {"PROTEIN": ["Histone"], "ASTRONOMICAL_OBJECT": [], "ENZYME": ["DNA polymerase", "RNA polymerase"], "MISC": ["cytoplasm"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ASTRONOMICAL_OBJECT, ENZYME, MISC\n\nText:\nHistone s, gene regulatory proteins, DNA polymerase and RNA polymerase s, and other substances essential for nuclear activities must be imported from the cytoplasm.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ASTRONOMICAL_OBJECT, ENZYME, MISC\n\nText:\nHistone s, gene regulatory proteins, DNA polymerase and RNA polymerase s, and other substances essential for nuclear activities must be imported from the cytoplasm.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_226", "text": "The recurring characters of Adele Webber (Loretta Devine), Finn Dandrige (Chris O'Donnell), Dr. Ellis Grey (Kate Burton) and Olivia Harper (Sarah Utterback) were portrayed with guest star billing.", "requested_types": ["ENZYME", "PERSON"], "distractors": ["ENZYME"], "gold": {"ENZYME": [], "PERSON": ["Adele Webber", "Loretta Devine", "Finn Dandrige", "Chris O'Donnell", "Dr. Ellis Grey", "Kate Burton", "Olivia Harper", "Sarah Utterback"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, PERSON\n\nText:\nThe recurring characters of Adele Webber (Loretta Devine), Finn Dandrige (Chris O'Donnell), Dr. Ellis Grey (Kate Burton) and Olivia Harper (Sarah Utterback) were portrayed with guest star billing.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, PERSON\n\nText:\nThe recurring characters of Adele Webber (Loretta Devine), Finn Dandrige (Chris O'Donnell), Dr. Ellis Grey (Kate Burton) and Olivia Harper (Sarah Utterback) were portrayed with guest star billing.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_227", "text": "(The previously discovered 45 Eugenia, 54 Alexandra, and 64 Angelina had nevertheless also been given non-classical names; 64 Angelina had also been discovered by Tempel, but its name stood despite similar protests.", "requested_types": ["LOCATION", "SCIENTIST", "ASTRONOMICAL_OBJECT"], "distractors": ["LOCATION"], "gold": {"LOCATION": [], "SCIENTIST": ["Tempel"], "ASTRONOMICAL_OBJECT": ["45 Eugenia", "54 Alexandra", "64 Angelina"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, SCIENTIST, ASTRONOMICAL_OBJECT\n\nText:\n(The previously discovered 45 Eugenia, 54 Alexandra, and 64 Angelina had nevertheless also been given non-classical names; 64 Angelina had also been discovered by Tempel, but its name stood despite similar protests.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, SCIENTIST, ASTRONOMICAL_OBJECT\n\nText:\n(The previously discovered 45 Eugenia, 54 Alexandra, and 64 Angelina had nevertheless also been given non-classical names; 64 Angelina had also been discovered by Tempel, but its name stood despite similar protests.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_228", "text": "Against a Crooked Sky is a 1975 American Western film directed by Earl Bellamy and starring Richard Boone, Stewart Petersen, and Henry Wilcoxon.", "requested_types": ["MISC", "COUNTRY", "THEORY", "PERSON"], "distractors": ["COUNTRY", "THEORY"], "gold": {"MISC": ["Crooked Sky", "American", "Western"], "COUNTRY": [], "THEORY": [], "PERSON": ["Earl Bellamy", "Richard Boone", "Stewart Petersen", "Henry Wilcoxon"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, COUNTRY, THEORY, PERSON\n\nText:\nAgainst a Crooked Sky is a 1975 American Western film directed by Earl Bellamy and starring Richard Boone, Stewart Petersen, and Henry Wilcoxon.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, COUNTRY, THEORY, PERSON\n\nText:\nAgainst a Crooked Sky is a 1975 American Western film directed by Earl Bellamy and starring Richard Boone, Stewart Petersen, and Henry Wilcoxon.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_229", "text": "Other alumni include scientists who were distinguished with the highest honours in their respective fields, amongst them Fields Medal, Pritzker Prize and Turing Award winners.", "requested_types": ["AWARD", "THEORY", "COUNTRY"], "distractors": ["THEORY", "COUNTRY"], "gold": {"AWARD": ["Fields Medal", "Pritzker Prize", "Turing Award"], "THEORY": [], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, THEORY, COUNTRY\n\nText:\nOther alumni include scientists who were distinguished with the highest honours in their respective fields, amongst them Fields Medal, Pritzker Prize and Turing Award winners.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, THEORY, COUNTRY\n\nText:\nOther alumni include scientists who were distinguished with the highest honours in their respective fields, amongst them Fields Medal, Pritzker Prize and Turing Award winners.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_230", "text": "His research in cognitive psychology has won the Early Career Award (1984) and Boyd McCandless Award (1986) from the American Psychological Association, the Troland Research Award (1993) from the National Academy of Sciences, the Henry Dale Prize (2004) from the Royal Institution of Great Britain, and the George Miller Prize (2010) from the Cognitive Neuroscience Society.", "requested_types": ["CHEMICAL_COMPOUND", "AWARD", "SCIENTIST", "ORGANISATION", "PERSON", "DISCIPLINE"], "distractors": ["SCIENTIST", "PERSON", "CHEMICAL_COMPOUND"], "gold": {"CHEMICAL_COMPOUND": [], "AWARD": ["Early Career Award", "Boyd McCandless Award", "Troland Research Award", "Henry Dale Prize", "George Miller Prize"], "SCIENTIST": [], "ORGANISATION": ["American Psychological Association", "National Academy of Sciences", "Royal Institution of Great Britain", "Cognitive Neuroscience Society"], "PERSON": [], "DISCIPLINE": ["cognitive psychology"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, AWARD, SCIENTIST, ORGANISATION, PERSON, DISCIPLINE\n\nText:\nHis research in cognitive psychology has won the Early Career Award (1984) and Boyd McCandless Award (1986) from the American Psychological Association, the Troland Research Award (1993) from the National Academy of Sciences, the Henry Dale Prize (2004) from the Royal Institution of Great Britain, and the George Miller Prize (2010) from the Cognitive Neuroscience Society.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, AWARD, SCIENTIST, ORGANISATION, PERSON, DISCIPLINE\n\nText:\nHis research in cognitive psychology has won the Early Career Award (1984) and Boyd McCandless Award (1986) from the American Psychological Association, the Troland Research Award (1993) from the National Academy of Sciences, the Henry Dale Prize (2004) from the Royal Institution of Great Britain, and the George Miller Prize (2010) from the Cognitive Neuroscience Society.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_231", "text": "Her children were Jupiter, Neptune, Pluto, Juno, Ceres, and Vesta.", "requested_types": ["ASTRONOMICAL_OBJECT", "UNIVERSITY"], "distractors": ["UNIVERSITY"], "gold": {"ASTRONOMICAL_OBJECT": ["Jupiter", "Neptune", "Pluto", "Juno", "Ceres", "Vesta"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, UNIVERSITY\n\nText:\nHer children were Jupiter, Neptune, Pluto, Juno, Ceres, and Vesta.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, UNIVERSITY\n\nText:\nHer children were Jupiter, Neptune, Pluto, Juno, Ceres, and Vesta.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_232", "text": "The episode was written by series creator Shonda Rhimes and directed by co-executive producer Peter Horton.", "requested_types": ["PERSON", "LOCATION", "EVENT"], "distractors": ["EVENT", "LOCATION"], "gold": {"PERSON": ["Shonda Rhimes", "Peter Horton"], "LOCATION": [], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, EVENT\n\nText:\nThe episode was written by series creator Shonda Rhimes and directed by co-executive producer Peter Horton.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, EVENT\n\nText:\nThe episode was written by series creator Shonda Rhimes and directed by co-executive producer Peter Horton.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_233", "text": "Cohen received the Louisa Gross Horwitz Prize from Columbia University together with Rita Levi-Montalcini in 1983, the Nobel Prize in Physiology or Medicine in 1986 for the isolation of nerve growth factor and the discovery of epidermal growth factor and the National Medal of Science in 1986.", "requested_types": ["SCIENTIST", "ASTRONOMICAL_OBJECT", "AWARD", "UNIVERSITY", "MISC", "ENZYME"], "distractors": ["ENZYME", "ASTRONOMICAL_OBJECT"], "gold": {"SCIENTIST": ["Cohen", "Rita Levi-Montalcini"], "ASTRONOMICAL_OBJECT": [], "AWARD": ["Louisa Gross Horwitz Prize", "Nobel Prize in Physiology or Medicine", "National Medal of Science"], "UNIVERSITY": ["Columbia University"], "MISC": ["nerve growth factor", "epidermal growth factor"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ASTRONOMICAL_OBJECT, AWARD, UNIVERSITY, MISC, ENZYME\n\nText:\nCohen received the Louisa Gross Horwitz Prize from Columbia University together with Rita Levi-Montalcini in 1983, the Nobel Prize in Physiology or Medicine in 1986 for the isolation of nerve growth factor and the discovery of epidermal growth factor and the National Medal of Science in 1986.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ASTRONOMICAL_OBJECT, AWARD, UNIVERSITY, MISC, ENZYME\n\nText:\nCohen received the Louisa Gross Horwitz Prize from Columbia University together with Rita Levi-Montalcini in 1983, the Nobel Prize in Physiology or Medicine in 1986 for the isolation of nerve growth factor and the discovery of epidermal growth factor and the National Medal of Science in 1986.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_234", "text": "He was a member of the Royal Swedish Academy of Sciences from 1836 and a foreign member of the Royal Society of London from 1837.", "requested_types": ["PERSON", "CHEMICAL_COMPOUND", "CHEMICAL_ELEMENT", "ORGANISATION"], "distractors": ["CHEMICAL_COMPOUND", "PERSON", "CHEMICAL_ELEMENT"], "gold": {"PERSON": [], "CHEMICAL_COMPOUND": [], "CHEMICAL_ELEMENT": [], "ORGANISATION": ["Royal Swedish Academy of Sciences", "Royal Society of London"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, CHEMICAL_COMPOUND, CHEMICAL_ELEMENT, ORGANISATION\n\nText:\nHe was a member of the Royal Swedish Academy of Sciences from 1836 and a foreign member of the Royal Society of London from 1837.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, CHEMICAL_COMPOUND, CHEMICAL_ELEMENT, ORGANISATION\n\nText:\nHe was a member of the Royal Swedish Academy of Sciences from 1836 and a foreign member of the Royal Society of London from 1837.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_235", "text": "During its ten seasons, Smallville won awards ranging from Primetime Emmy Award to Teen Choice Awards.", "requested_types": ["PERSON", "CHEMICAL_ELEMENT", "AWARD", "ASTRONOMICAL_OBJECT", "EVENT"], "distractors": ["EVENT", "ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT"], "gold": {"PERSON": ["Smallville"], "CHEMICAL_ELEMENT": [], "AWARD": ["Primetime Emmy Award", "Teen Choice Awards"], "ASTRONOMICAL_OBJECT": [], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, CHEMICAL_ELEMENT, AWARD, ASTRONOMICAL_OBJECT, EVENT\n\nText:\nDuring its ten seasons, Smallville won awards ranging from Primetime Emmy Award to Teen Choice Awards.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, CHEMICAL_ELEMENT, AWARD, ASTRONOMICAL_OBJECT, EVENT\n\nText:\nDuring its ten seasons, Smallville won awards ranging from Primetime Emmy Award to Teen Choice Awards.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_236", "text": "Michalak's estimate depends on the masses of 19 Fortuna, 29 Amphitrite, and 16 Psyche; thus this mass was obtained assuming an incomplete dynamical model.", "requested_types": ["PERSON", "EVENT", "ASTRONOMICAL_OBJECT", "AWARD"], "distractors": ["AWARD", "EVENT"], "gold": {"PERSON": ["Michalak"], "EVENT": [], "ASTRONOMICAL_OBJECT": ["19 Fortuna", "29 Amphitrite", "16 Psyche"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, EVENT, ASTRONOMICAL_OBJECT, AWARD\n\nText:\nMichalak's estimate depends on the masses of 19 Fortuna, 29 Amphitrite, and 16 Psyche; thus this mass was obtained assuming an incomplete dynamical model.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, EVENT, ASTRONOMICAL_OBJECT, AWARD\n\nText:\nMichalak's estimate depends on the masses of 19 Fortuna, 29 Amphitrite, and 16 Psyche; thus this mass was obtained assuming an incomplete dynamical model.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_237", "text": "Such covalent substances are usually gases, for example, HCl, Sulfur dioxide, Carbon dioxide, and Methane.", "requested_types": ["ENZYME", "CHEMICAL_COMPOUND"], "distractors": ["ENZYME"], "gold": {"ENZYME": [], "CHEMICAL_COMPOUND": ["HCl", "Sulfur dioxide", "Carbon dioxide", "Methane"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, CHEMICAL_COMPOUND\n\nText:\nSuch covalent substances are usually gases, for example, HCl, Sulfur dioxide, Carbon dioxide, and Methane.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, CHEMICAL_COMPOUND\n\nText:\nSuch covalent substances are usually gases, for example, HCl, Sulfur dioxide, Carbon dioxide, and Methane.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_238", "text": "Directors of the botanical garden from this period included Karl Julius Perleb, Fridolin Karl Leopold Spenner, Alexander Braun, Carl Wilhelm von Nägeli, Heinrich Anton de Bary and Julius von Sachs.", "requested_types": ["SCIENTIST", "CHEMICAL_COMPOUND", "MISC"], "distractors": ["CHEMICAL_COMPOUND"], "gold": {"SCIENTIST": ["Karl Julius Perleb", "Fridolin Karl Leopold Spenner", "Alexander Braun", "Carl Wilhelm von Nägeli", "Heinrich Anton de Bary", "Julius von Sachs"], "CHEMICAL_COMPOUND": [], "MISC": ["botanical garden"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_COMPOUND, MISC\n\nText:\nDirectors of the botanical garden from this period included Karl Julius Perleb, Fridolin Karl Leopold Spenner, Alexander Braun, Carl Wilhelm von Nägeli, Heinrich Anton de Bary and Julius von Sachs.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_COMPOUND, MISC\n\nText:\nDirectors of the botanical garden from this period included Karl Julius Perleb, Fridolin Karl Leopold Spenner, Alexander Braun, Carl Wilhelm von Nägeli, Heinrich Anton de Bary and Julius von Sachs.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_239", "text": "The most influential people were Erich Schumann, Abraham Esau, Walther Gerlach, and Kurt Diebner.", "requested_types": ["PROTEIN", "ORGANISATION", "CHEMICAL_ELEMENT", "PERSON"], "distractors": ["ORGANISATION", "PROTEIN", "CHEMICAL_ELEMENT"], "gold": {"PROTEIN": [], "ORGANISATION": [], "CHEMICAL_ELEMENT": [], "PERSON": ["Erich Schumann", "Abraham Esau", "Walther Gerlach", "Kurt Diebner"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ORGANISATION, CHEMICAL_ELEMENT, PERSON\n\nText:\nThe most influential people were Erich Schumann, Abraham Esau, Walther Gerlach, and Kurt Diebner.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ORGANISATION, CHEMICAL_ELEMENT, PERSON\n\nText:\nThe most influential people were Erich Schumann, Abraham Esau, Walther Gerlach, and Kurt Diebner.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_240", "text": "Paleontologists - Mary Anning - Robert T. Bakker - Barnum Brown - William Buckland - Edward Drinker Cope - Jack Horner - Gideon Mantell - Othniel Charles Marsh - John Ostrom - Dong Zhiming br /", "requested_types": ["DISCIPLINE", "CHEMICAL_ELEMENT", "SCIENTIST"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"DISCIPLINE": ["Paleontologists"], "CHEMICAL_ELEMENT": [], "SCIENTIST": ["Mary Anning", "Robert T. Bakker", "Barnum Brown", "William Buckland", "Edward Drinker Cope", "Jack Horner", "Gideon Mantell", "Othniel Charles Marsh", "John Ostrom", "Dong Zhiming"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, CHEMICAL_ELEMENT, SCIENTIST\n\nText:\nPaleontologists - Mary Anning - Robert T. Bakker - Barnum Brown - William Buckland - Edward Drinker Cope - Jack Horner - Gideon Mantell - Othniel Charles Marsh - John Ostrom - Dong Zhiming br /\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, CHEMICAL_ELEMENT, SCIENTIST\n\nText:\nPaleontologists - Mary Anning - Robert T. Bakker - Barnum Brown - William Buckland - Edward Drinker Cope - Jack Horner - Gideon Mantell - Othniel Charles Marsh - John Ostrom - Dong Zhiming br /\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_241", "text": "In May 2008, two new exoplanets of Jupiter size, CoRoT-4b and CoRoT-5b, as well as an unknown massive celestial object, CoRoT-3b, were announced by ESA.", "requested_types": ["ORGANISATION", "EVENT", "ASTRONOMICAL_OBJECT", "LOCATION"], "distractors": ["LOCATION", "EVENT"], "gold": {"ORGANISATION": ["ESA"], "EVENT": [], "ASTRONOMICAL_OBJECT": ["Jupiter", "CoRoT-4b", "CoRoT-5b", "CoRoT-3b"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, EVENT, ASTRONOMICAL_OBJECT, LOCATION\n\nText:\nIn May 2008, two new exoplanets of Jupiter size, CoRoT-4b and CoRoT-5b, as well as an unknown massive celestial object, CoRoT-3b, were announced by ESA.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, EVENT, ASTRONOMICAL_OBJECT, LOCATION\n\nText:\nIn May 2008, two new exoplanets of Jupiter size, CoRoT-4b and CoRoT-5b, as well as an unknown massive celestial object, CoRoT-3b, were announced by ESA.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_242", "text": "The Anti-Defamation League and the Southern Poverty Law Center assert that Pierce created cosmotheism in order to acquire tax-exempt status for the National Alliance after he had failed to do so earlier, and the SPLC refers to it as a bogus religion.", "requested_types": ["ORGANISATION", "THEORY", "PERSON", "EVENT", "PROTEIN"], "distractors": ["EVENT", "PROTEIN"], "gold": {"ORGANISATION": ["Anti-Defamation League", "Southern Poverty Law Center", "National Alliance", "SPLC"], "THEORY": ["cosmotheism"], "PERSON": ["Pierce"], "EVENT": [], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, THEORY, PERSON, EVENT, PROTEIN\n\nText:\nThe Anti-Defamation League and the Southern Poverty Law Center assert that Pierce created cosmotheism in order to acquire tax-exempt status for the National Alliance after he had failed to do so earlier, and the SPLC refers to it as a bogus religion.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, THEORY, PERSON, EVENT, PROTEIN\n\nText:\nThe Anti-Defamation League and the Southern Poverty Law Center assert that Pierce created cosmotheism in order to acquire tax-exempt status for the National Alliance after he had failed to do so earlier, and the SPLC refers to it as a bogus religion.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_243", "text": "In the episode, Sarah Jane Smith (Elisabeth Sladen) and Martha Jones (Freema Agyeman), former travelling companions of the time-travelling Tenth Doctor (David Tennant), prepare to counter the Dalek s' universe-destroying Reality Bomb with devastating weapons of their own.", "requested_types": ["MISC", "PERSON", "SCIENTIST"], "distractors": ["SCIENTIST"], "gold": {"MISC": ["time-travelling", "Dalek"], "PERSON": ["Sarah Jane Smith", "Elisabeth Sladen", "Martha Jones", "Freema Agyeman", "Tenth Doctor", "David Tennant"], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PERSON, SCIENTIST\n\nText:\nIn the episode, Sarah Jane Smith (Elisabeth Sladen) and Martha Jones (Freema Agyeman), former travelling companions of the time-travelling Tenth Doctor (David Tennant), prepare to counter the Dalek s' universe-destroying Reality Bomb with devastating weapons of their own.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PERSON, SCIENTIST\n\nText:\nIn the episode, Sarah Jane Smith (Elisabeth Sladen) and Martha Jones (Freema Agyeman), former travelling companions of the time-travelling Tenth Doctor (David Tennant), prepare to counter the Dalek s' universe-destroying Reality Bomb with devastating weapons of their own.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_244", "text": "Other studied asteroids included 15 Eunomia, 43 Ariadne, 44 Nysa, and 624 Hektor.", "requested_types": ["ASTRONOMICAL_OBJECT", "EVENT"], "distractors": ["EVENT"], "gold": {"ASTRONOMICAL_OBJECT": ["15 Eunomia", "43 Ariadne", "44 Nysa", "624 Hektor"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, EVENT\n\nText:\nOther studied asteroids included 15 Eunomia, 43 Ariadne, 44 Nysa, and 624 Hektor.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, EVENT\n\nText:\nOther studied asteroids included 15 Eunomia, 43 Ariadne, 44 Nysa, and 624 Hektor.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_245", "text": "On August 27, Mercury and Venus were in conjunction, followed by a conjunction of Venus and Jupiter, meaning that the three planets were very close together in the evening sky.", "requested_types": ["MISC", "ASTRONOMICAL_OBJECT", "UNIVERSITY"], "distractors": ["MISC", "UNIVERSITY"], "gold": {"MISC": [], "ASTRONOMICAL_OBJECT": ["Mercury", "Venus", "Jupiter"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ASTRONOMICAL_OBJECT, UNIVERSITY\n\nText:\nOn August 27, Mercury and Venus were in conjunction, followed by a conjunction of Venus and Jupiter, meaning that the three planets were very close together in the evening sky.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ASTRONOMICAL_OBJECT, UNIVERSITY\n\nText:\nOn August 27, Mercury and Venus were in conjunction, followed by a conjunction of Venus and Jupiter, meaning that the three planets were very close together in the evening sky.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_246", "text": "Audiences in Monte Carlo, Warsaw and Buenos Aires also heard Caruso sing during this pivotal phase of his career and, in 1899-1900, he appeared before the tsar and the Russian aristocracy at the Mariinsky Theatre in Saint Petersburg and the Bolshoi Theatre in Moscow as part of a touring company of first-class Italian singers.", "requested_types": ["PERSON", "MISC", "LOCATION", "SCIENTIST"], "distractors": ["SCIENTIST"], "gold": {"PERSON": ["Caruso"], "MISC": ["Russian", "Italian"], "LOCATION": ["Monte Carlo", "Warsaw", "Buenos Aires", "Mariinsky Theatre", "Bolshoi Theatre", "Moscow"], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, MISC, LOCATION, SCIENTIST\n\nText:\nAudiences in Monte Carlo, Warsaw and Buenos Aires also heard Caruso sing during this pivotal phase of his career and, in 1899-1900, he appeared before the tsar and the Russian aristocracy at the Mariinsky Theatre in Saint Petersburg and the Bolshoi Theatre in Moscow as part of a touring company of first-class Italian singers.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, MISC, LOCATION, SCIENTIST\n\nText:\nAudiences in Monte Carlo, Warsaw and Buenos Aires also heard Caruso sing during this pivotal phase of his career and, in 1899-1900, he appeared before the tsar and the Russian aristocracy at the Mariinsky Theatre in Saint Petersburg and the Bolshoi Theatre in Moscow as part of a touring company of first-class Italian singers.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_247", "text": "Another small epoch seen as a golden age is the unification of electricity, magnetism, and optics because of 19th century notables, including Michael Faraday, James Clerk Maxwell, and others.", "requested_types": ["PERSON", "SCIENTIST", "DISCIPLINE"], "distractors": ["PERSON"], "gold": {"PERSON": [], "SCIENTIST": ["Michael Faraday", "James Clerk Maxwell"], "DISCIPLINE": ["electricity", "magnetism", "optics"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, SCIENTIST, DISCIPLINE\n\nText:\nAnother small epoch seen as a golden age is the unification of electricity, magnetism, and optics because of 19th century notables, including Michael Faraday, James Clerk Maxwell, and others.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, SCIENTIST, DISCIPLINE\n\nText:\nAnother small epoch seen as a golden age is the unification of electricity, magnetism, and optics because of 19th century notables, including Michael Faraday, James Clerk Maxwell, and others.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_248", "text": "This culminates in the translocation of the NF-κB transcription factor Relish, leading to production of antimicrobial peptides and other effectors.", "requested_types": ["PROTEIN", "CHEMICAL_COMPOUND", "ACADEMIC_JOURNAL", "MISC"], "distractors": ["ACADEMIC_JOURNAL", "CHEMICAL_COMPOUND"], "gold": {"PROTEIN": ["NF-κB"], "CHEMICAL_COMPOUND": [], "ACADEMIC_JOURNAL": [], "MISC": ["Relish", "antimicrobial peptides"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_COMPOUND, ACADEMIC_JOURNAL, MISC\n\nText:\nThis culminates in the translocation of the NF-κB transcription factor Relish, leading to production of antimicrobial peptides and other effectors.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_COMPOUND, ACADEMIC_JOURNAL, MISC\n\nText:\nThis culminates in the translocation of the NF-κB transcription factor Relish, leading to production of antimicrobial peptides and other effectors.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_249", "text": "Like anandamide, both compounds activated the Cannabinoid receptor.", "requested_types": ["PROTEIN", "COUNTRY", "CHEMICAL_COMPOUND", "EVENT"], "distractors": ["COUNTRY", "EVENT"], "gold": {"PROTEIN": ["Cannabinoid receptor"], "COUNTRY": [], "CHEMICAL_COMPOUND": ["anandamide"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, COUNTRY, CHEMICAL_COMPOUND, EVENT\n\nText:\nLike anandamide, both compounds activated the Cannabinoid receptor.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, COUNTRY, CHEMICAL_COMPOUND, EVENT\n\nText:\nLike anandamide, both compounds activated the Cannabinoid receptor.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_250", "text": "The first one, Annales de physique, was latterly published by EDP Sciences under the same name up to 2009, when it became integrated in the European Physical Journal series as the European Physical Journal H.", "requested_types": ["COUNTRY", "ORGANISATION", "ACADEMIC_JOURNAL"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "ORGANISATION": ["EDP Sciences"], "ACADEMIC_JOURNAL": ["Annales de physique", "European Physical Journal", "European Physical Journal H"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ORGANISATION, ACADEMIC_JOURNAL\n\nText:\nThe first one, Annales de physique, was latterly published by EDP Sciences under the same name up to 2009, when it became integrated in the European Physical Journal series as the European Physical Journal H.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ORGANISATION, ACADEMIC_JOURNAL\n\nText:\nThe first one, Annales de physique, was latterly published by EDP Sciences under the same name up to 2009, when it became integrated in the European Physical Journal series as the European Physical Journal H.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_251", "text": "Many scientists such as James Clerk Maxwell, Oliver Heaviside and Heinrich Hertz unsuccessfully attempted to solve these problems by incorporating either Fresnel or Stokes' theories into Maxwell's new electromagnetic laws.", "requested_types": ["PROTEIN", "SCIENTIST", "ASTRONOMICAL_OBJECT", "THEORY", "ACADEMIC_JOURNAL"], "distractors": ["ACADEMIC_JOURNAL", "ASTRONOMICAL_OBJECT", "PROTEIN"], "gold": {"PROTEIN": [], "SCIENTIST": ["James Clerk Maxwell", "Oliver Heaviside", "Heinrich Hertz", "Fresnel", "Stokes"], "ASTRONOMICAL_OBJECT": [], "THEORY": ["Maxwell's new electromagnetic laws"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST, ASTRONOMICAL_OBJECT, THEORY, ACADEMIC_JOURNAL\n\nText:\nMany scientists such as James Clerk Maxwell, Oliver Heaviside and Heinrich Hertz unsuccessfully attempted to solve these problems by incorporating either Fresnel or Stokes' theories into Maxwell's new electromagnetic laws.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST, ASTRONOMICAL_OBJECT, THEORY, ACADEMIC_JOURNAL\n\nText:\nMany scientists such as James Clerk Maxwell, Oliver Heaviside and Heinrich Hertz unsuccessfully attempted to solve these problems by incorporating either Fresnel or Stokes' theories into Maxwell's new electromagnetic laws.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_252", "text": "Furry made important contributions to the early development of Quantum Field Theory with J. Robert Oppenheimer, Vladimir Fock, and others.", "requested_types": ["PROTEIN", "SCIENTIST", "THEORY"], "distractors": ["PROTEIN"], "gold": {"PROTEIN": [], "SCIENTIST": ["Furry", "J. Robert Oppenheimer", "Vladimir Fock"], "THEORY": ["Quantum Field Theory"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST, THEORY\n\nText:\nFurry made important contributions to the early development of Quantum Field Theory with J. Robert Oppenheimer, Vladimir Fock, and others.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST, THEORY\n\nText:\nFurry made important contributions to the early development of Quantum Field Theory with J. Robert Oppenheimer, Vladimir Fock, and others.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_253", "text": "While Caruso sang at such venues as La Scala in Milan, the Royal Opera House, in London, the Mariinsky Theatre in Saint Petersburg, and the Teatro Colón in Buenos Aires, he appeared most often at the Metropolitan Opera in New York City where he was the leading tenor for 18 consecutive seasons.", "requested_types": ["PERSON", "LOCATION", "PROTEIN", "SCIENTIST"], "distractors": ["PROTEIN", "SCIENTIST"], "gold": {"PERSON": ["Caruso"], "LOCATION": ["La Scala", "Milan", "Royal Opera House", "London", "Mariinsky Theatre", "Saint Petersburg", "Teatro Colón", "Buenos Aires", "Metropolitan Opera", "New York City"], "PROTEIN": [], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, PROTEIN, SCIENTIST\n\nText:\nWhile Caruso sang at such venues as La Scala in Milan, the Royal Opera House, in London, the Mariinsky Theatre in Saint Petersburg, and the Teatro Colón in Buenos Aires, he appeared most often at the Metropolitan Opera in New York City where he was the leading tenor for 18 consecutive seasons.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, PROTEIN, SCIENTIST\n\nText:\nWhile Caruso sang at such venues as La Scala in Milan, the Royal Opera House, in London, the Mariinsky Theatre in Saint Petersburg, and the Teatro Colón in Buenos Aires, he appeared most often at the Metropolitan Opera in New York City where he was the leading tenor for 18 consecutive seasons.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_254", "text": "These are: Birdlife International, the IUCN, the International Water Management Institute, Wetlands International and the World Wide Fund for Nature.", "requested_types": ["EVENT", "ORGANISATION"], "distractors": ["EVENT"], "gold": {"EVENT": [], "ORGANISATION": ["Birdlife International", "IUCN", "International Water Management Institute", "Wetlands International", "World Wide Fund for Nature"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, ORGANISATION\n\nText:\nThese are: Birdlife International, the IUCN, the International Water Management Institute, Wetlands International and the World Wide Fund for Nature.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, ORGANISATION\n\nText:\nThese are: Birdlife International, the IUCN, the International Water Management Institute, Wetlands International and the World Wide Fund for Nature.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_255", "text": "2016 he has been awarded the Kavli Prize in Nanoscience together with Gerd Binnig and Calvin Quate for the Scanning Force Microscope.", "requested_types": ["MISC", "SCIENTIST", "DISCIPLINE", "PERSON", "AWARD", "THEORY"], "distractors": ["THEORY", "PERSON"], "gold": {"MISC": ["Scanning Force Microscope"], "SCIENTIST": ["Gerd Binnig", "Calvin Quate"], "DISCIPLINE": ["Nanoscience"], "PERSON": [], "AWARD": ["Kavli Prize"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST, DISCIPLINE, PERSON, AWARD, THEORY\n\nText:\n2016 he has been awarded the Kavli Prize in Nanoscience together with Gerd Binnig and Calvin Quate for the Scanning Force Microscope.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST, DISCIPLINE, PERSON, AWARD, THEORY\n\nText:\n2016 he has been awarded the Kavli Prize in Nanoscience together with Gerd Binnig and Calvin Quate for the Scanning Force Microscope.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_256", "text": "A nucleosome consists of 2 sets of 4 histones: H2A, H2B, Histone H3, and Histone H4.", "requested_types": ["PROTEIN", "ENZYME", "ACADEMIC_JOURNAL"], "distractors": ["ACADEMIC_JOURNAL", "ENZYME"], "gold": {"PROTEIN": ["H2A", "H2B", "Histone H3", "Histone H4"], "ENZYME": [], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ENZYME, ACADEMIC_JOURNAL\n\nText:\nA nucleosome consists of 2 sets of 4 histones: H2A, H2B, Histone H3, and Histone H4.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ENZYME, ACADEMIC_JOURNAL\n\nText:\nA nucleosome consists of 2 sets of 4 histones: H2A, H2B, Histone H3, and Histone H4.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_257", "text": "In 1978, he was also heavily influenced by the visit of Simon van der Meer and Carlo Rubbia to the laboratory, speaking on the possibilities of stochastic phase space cooling of antiproton beams and the exciting possibilities with proton-antiproton collisions.", "requested_types": ["PERSON", "SCIENTIST", "MISC", "UNIVERSITY"], "distractors": ["PERSON", "UNIVERSITY"], "gold": {"PERSON": [], "SCIENTIST": ["Simon van der Meer", "Carlo Rubbia"], "MISC": ["stochastic phase space cooling", "antiproton", "proton-antiproton collisions"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, SCIENTIST, MISC, UNIVERSITY\n\nText:\nIn 1978, he was also heavily influenced by the visit of Simon van der Meer and Carlo Rubbia to the laboratory, speaking on the possibilities of stochastic phase space cooling of antiproton beams and the exciting possibilities with proton-antiproton collisions.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, SCIENTIST, MISC, UNIVERSITY\n\nText:\nIn 1978, he was also heavily influenced by the visit of Simon van der Meer and Carlo Rubbia to the laboratory, speaking on the possibilities of stochastic phase space cooling of antiproton beams and the exciting possibilities with proton-antiproton collisions.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_258", "text": "Botryosphaeran reduces hepatic steatosis and dyslipidaemia, and glucose intolerance in diet-induced obese rats through activation of AMP-activated protein kinase (AMPK) and the expression of the Forkhead transcription factor, FOXO3a, in adipose tissue.", "requested_types": ["MISC", "PROTEIN", "THEORY", "ENZYME", "PERSON", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND", "PERSON", "THEORY"], "gold": {"MISC": ["Botryosphaeran", "hepatic steatosis", "dyslipidaemia", "glucose intolerance", "Forkhead transcription factor", "adipose tissue"], "PROTEIN": ["FOXO3a"], "THEORY": [], "ENZYME": ["AMP-activated protein kinase", "AMPK"], "PERSON": [], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, THEORY, ENZYME, PERSON, CHEMICAL_COMPOUND\n\nText:\nBotryosphaeran reduces hepatic steatosis and dyslipidaemia, and glucose intolerance in diet-induced obese rats through activation of AMP-activated protein kinase (AMPK) and the expression of the Forkhead transcription factor, FOXO3a, in adipose tissue.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, THEORY, ENZYME, PERSON, CHEMICAL_COMPOUND\n\nText:\nBotryosphaeran reduces hepatic steatosis and dyslipidaemia, and glucose intolerance in diet-induced obese rats through activation of AMP-activated protein kinase (AMPK) and the expression of the Forkhead transcription factor, FOXO3a, in adipose tissue.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_259", "text": "A 1990 Michael Winner film Bullseye!, starring Michael Caine and Roger Moore, referenced the Fleischmann and Pons experiment.", "requested_types": ["MISC", "PERSON", "ACADEMIC_JOURNAL", "ASTRONOMICAL_OBJECT"], "distractors": ["ACADEMIC_JOURNAL", "ASTRONOMICAL_OBJECT"], "gold": {"MISC": ["Bullseye!", "Fleischmann and Pons experiment"], "PERSON": ["Michael Winner", "Michael Caine", "Roger Moore"], "ACADEMIC_JOURNAL": [], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PERSON, ACADEMIC_JOURNAL, ASTRONOMICAL_OBJECT\n\nText:\nA 1990 Michael Winner film Bullseye!, starring Michael Caine and Roger Moore, referenced the Fleischmann and Pons experiment.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PERSON, ACADEMIC_JOURNAL, ASTRONOMICAL_OBJECT\n\nText:\nA 1990 Michael Winner film Bullseye!, starring Michael Caine and Roger Moore, referenced the Fleischmann and Pons experiment.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_260", "text": "She also played at 1990 Asian Games, 1991, 1993 and 1995 AFC Championship.", "requested_types": ["PROTEIN", "COUNTRY", "UNIVERSITY", "EVENT"], "distractors": ["PROTEIN", "UNIVERSITY", "COUNTRY"], "gold": {"PROTEIN": [], "COUNTRY": [], "UNIVERSITY": [], "EVENT": ["Asian Games", "AFC Championship"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, COUNTRY, UNIVERSITY, EVENT\n\nText:\nShe also played at 1990 Asian Games, 1991, 1993 and 1995 AFC Championship.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, COUNTRY, UNIVERSITY, EVENT\n\nText:\nShe also played at 1990 Asian Games, 1991, 1993 and 1995 AFC Championship.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_261", "text": "In the Greek-influenced tradition, Neptune is the brother of Jupiter and Pluto; the brothers preside over the realms of Heaven, the earthly world, and the Underworld .About the relationship of the lord of our earthly world with water (s) Bloch, p.", "requested_types": ["PERSON", "ENZYME", "MISC"], "distractors": ["ENZYME"], "gold": {"PERSON": ["Neptune", "Jupiter", "Pluto", "Bloch"], "ENZYME": [], "MISC": ["Greek-influenced"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ENZYME, MISC\n\nText:\nIn the Greek-influenced tradition, Neptune is the brother of Jupiter and Pluto; the brothers preside over the realms of Heaven, the earthly world, and the Underworld .About the relationship of the lord of our earthly world with water (s) Bloch, p.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ENZYME, MISC\n\nText:\nIn the Greek-influenced tradition, Neptune is the brother of Jupiter and Pluto; the brothers preside over the realms of Heaven, the earthly world, and the Underworld .About the relationship of the lord of our earthly world with water (s) Bloch, p.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_262", "text": "First subgroup composed by JAM-A, JAM-B, JAM-C has a PDZ domain binding motif type II at their C-termini, which interacts with the PDZ domain of ZO-1, AF-6, PAR-3 and MUPP1.", "requested_types": ["PROTEIN", "MISC", "EVENT", "ORGANISATION", "UNIVERSITY"], "distractors": ["ORGANISATION", "UNIVERSITY", "EVENT"], "gold": {"PROTEIN": ["JAM-A", "JAM-B", "JAM-C", "ZO-1", "AF-6", "PAR-3", "MUPP1"], "MISC": ["PDZ domain", "C-termini"], "EVENT": [], "ORGANISATION": [], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, EVENT, ORGANISATION, UNIVERSITY\n\nText:\nFirst subgroup composed by JAM-A, JAM-B, JAM-C has a PDZ domain binding motif type II at their C-termini, which interacts with the PDZ domain of ZO-1, AF-6, PAR-3 and MUPP1.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, EVENT, ORGANISATION, UNIVERSITY\n\nText:\nFirst subgroup composed by JAM-A, JAM-B, JAM-C has a PDZ domain binding motif type II at their C-termini, which interacts with the PDZ domain of ZO-1, AF-6, PAR-3 and MUPP1.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_263", "text": "(See Noyori asymmetric hydrogenation) Asymmetric hydrogenation of an alkene in the presence of ((S) - BINAP) Ru (Acetate) sub2 / sub is used for the commercial production of enantiomerically pure (97% ee) naproxen, used as an anti-inflammatory drug.", "requested_types": ["ORGANISATION", "THEORY", "CHEMICAL_ELEMENT", "MISC", "CHEMICAL_COMPOUND", "COUNTRY"], "distractors": ["ORGANISATION", "COUNTRY", "THEORY"], "gold": {"ORGANISATION": [], "THEORY": [], "CHEMICAL_ELEMENT": ["S", "Ru"], "MISC": ["Noyori asymmetric hydrogenation", "hydrogenation", "anti-inflammatory drug"], "CHEMICAL_COMPOUND": ["alkene", "BINAP", "Acetate", "ee", "naproxen"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, THEORY, CHEMICAL_ELEMENT, MISC, CHEMICAL_COMPOUND, COUNTRY\n\nText:\n(See Noyori asymmetric hydrogenation) Asymmetric hydrogenation of an alkene in the presence of ((S) - BINAP) Ru (Acetate) sub2 / sub is used for the commercial production of enantiomerically pure (97% ee) naproxen, used as an anti-inflammatory drug.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, THEORY, CHEMICAL_ELEMENT, MISC, CHEMICAL_COMPOUND, COUNTRY\n\nText:\n(See Noyori asymmetric hydrogenation) Asymmetric hydrogenation of an alkene in the presence of ((S) - BINAP) Ru (Acetate) sub2 / sub is used for the commercial production of enantiomerically pure (97% ee) naproxen, used as an anti-inflammatory drug.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_264", "text": "At the end of May, Mercury, Venus and Jupiter went through a series of conjunctions only a few days apart.", "requested_types": ["PROTEIN", "ASTRONOMICAL_OBJECT"], "distractors": ["PROTEIN"], "gold": {"PROTEIN": [], "ASTRONOMICAL_OBJECT": ["Mercury", "Venus", "Jupiter"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ASTRONOMICAL_OBJECT\n\nText:\nAt the end of May, Mercury, Venus and Jupiter went through a series of conjunctions only a few days apart.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ASTRONOMICAL_OBJECT\n\nText:\nAt the end of May, Mercury, Venus and Jupiter went through a series of conjunctions only a few days apart.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_265", "text": "There is some information about earthquakes in Aristotle 's Meteorology, in Naturalis Historia by Pliny the Elder, and in Strabo' s Geographica.", "requested_types": ["MISC", "CHEMICAL_ELEMENT", "ACADEMIC_JOURNAL", "PERSON", "SCIENTIST", "DISCIPLINE", "UNIVERSITY"], "distractors": ["ACADEMIC_JOURNAL", "UNIVERSITY", "CHEMICAL_ELEMENT"], "gold": {"MISC": ["Naturalis Historia"], "CHEMICAL_ELEMENT": [], "ACADEMIC_JOURNAL": [], "PERSON": ["Pliny the Elder"], "SCIENTIST": ["Aristotle", "Strabo"], "DISCIPLINE": ["Meteorology", "Geographica"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_ELEMENT, ACADEMIC_JOURNAL, PERSON, SCIENTIST, DISCIPLINE, UNIVERSITY\n\nText:\nThere is some information about earthquakes in Aristotle 's Meteorology, in Naturalis Historia by Pliny the Elder, and in Strabo' s Geographica.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_ELEMENT, ACADEMIC_JOURNAL, PERSON, SCIENTIST, DISCIPLINE, UNIVERSITY\n\nText:\nThere is some information about earthquakes in Aristotle 's Meteorology, in Naturalis Historia by Pliny the Elder, and in Strabo' s Geographica.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_266", "text": "Ernst Mayr remarks that the theory was hotly contested by some famous geneticists: William Bateson, Wilhelm Johannsen, Richard Goldschmidt and T.H. Morgan, all of a rather dogmatic turn of mind.", "requested_types": ["SCIENTIST", "ENZYME"], "distractors": ["ENZYME"], "gold": {"SCIENTIST": ["Ernst Mayr", "William Bateson", "Wilhelm Johannsen", "Richard Goldschmidt", "T.H. Morgan"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ENZYME\n\nText:\nErnst Mayr remarks that the theory was hotly contested by some famous geneticists: William Bateson, Wilhelm Johannsen, Richard Goldschmidt and T.H. Morgan, all of a rather dogmatic turn of mind.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ENZYME\n\nText:\nErnst Mayr remarks that the theory was hotly contested by some famous geneticists: William Bateson, Wilhelm Johannsen, Richard Goldschmidt and T.H. Morgan, all of a rather dogmatic turn of mind.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_267", "text": "He calculated the orbit s of the moons of Uranus and Saturn, obtaining the first value for Uranus' mass.", "requested_types": ["ORGANISATION", "ASTRONOMICAL_OBJECT", "ACADEMIC_JOURNAL"], "distractors": ["ACADEMIC_JOURNAL", "ORGANISATION"], "gold": {"ORGANISATION": [], "ASTRONOMICAL_OBJECT": ["Uranus", "Saturn"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ASTRONOMICAL_OBJECT, ACADEMIC_JOURNAL\n\nText:\nHe calculated the orbit s of the moons of Uranus and Saturn, obtaining the first value for Uranus' mass.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ASTRONOMICAL_OBJECT, ACADEMIC_JOURNAL\n\nText:\nHe calculated the orbit s of the moons of Uranus and Saturn, obtaining the first value for Uranus' mass.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_268", "text": "Some known inhibitors of the caveolae pathway are Filipin III, Genistein and Nystatin.", "requested_types": ["UNIVERSITY", "COUNTRY", "CHEMICAL_COMPOUND", "DISCIPLINE"], "distractors": ["COUNTRY", "UNIVERSITY", "DISCIPLINE"], "gold": {"UNIVERSITY": [], "COUNTRY": [], "CHEMICAL_COMPOUND": ["Filipin III", "Genistein", "Nystatin"], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, COUNTRY, CHEMICAL_COMPOUND, DISCIPLINE\n\nText:\nSome known inhibitors of the caveolae pathway are Filipin III, Genistein and Nystatin.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, COUNTRY, CHEMICAL_COMPOUND, DISCIPLINE\n\nText:\nSome known inhibitors of the caveolae pathway are Filipin III, Genistein and Nystatin.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_269", "text": "Three of the country's largest universities are located in central Thessaloniki: Aristotle University of Thessaloniki, the University of Macedonia and the International Hellenic University.", "requested_types": ["MISC", "UNIVERSITY", "LOCATION"], "distractors": ["MISC"], "gold": {"MISC": [], "UNIVERSITY": ["Aristotle University of Thessaloniki", "University of Macedonia", "International Hellenic University"], "LOCATION": ["central Thessaloniki"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, UNIVERSITY, LOCATION\n\nText:\nThree of the country's largest universities are located in central Thessaloniki: Aristotle University of Thessaloniki, the University of Macedonia and the International Hellenic University.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, UNIVERSITY, LOCATION\n\nText:\nThree of the country's largest universities are located in central Thessaloniki: Aristotle University of Thessaloniki, the University of Macedonia and the International Hellenic University.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_270", "text": "Walter Kohn, for his development of the density-functional theory, and John Pople, for his development of computational methods in quantum chemistry, received the 1998 Nobel Prize in Chemistry.", "requested_types": ["DISCIPLINE", "PROTEIN", "THEORY", "CHEMICAL_COMPOUND", "SCIENTIST", "AWARD"], "distractors": ["PROTEIN", "CHEMICAL_COMPOUND"], "gold": {"DISCIPLINE": ["quantum chemistry"], "PROTEIN": [], "THEORY": ["density-functional theory"], "CHEMICAL_COMPOUND": [], "SCIENTIST": ["Walter Kohn", "John Pople"], "AWARD": ["Nobel Prize in Chemistry"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, PROTEIN, THEORY, CHEMICAL_COMPOUND, SCIENTIST, AWARD\n\nText:\nWalter Kohn, for his development of the density-functional theory, and John Pople, for his development of computational methods in quantum chemistry, received the 1998 Nobel Prize in Chemistry.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, PROTEIN, THEORY, CHEMICAL_COMPOUND, SCIENTIST, AWARD\n\nText:\nWalter Kohn, for his development of the density-functional theory, and John Pople, for his development of computational methods in quantum chemistry, received the 1998 Nobel Prize in Chemistry.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_271", "text": "Max Planck, Albert Einstein, and Niels Bohr postulated the occurrence of energy in discrete quantities (quanta) in order to explain phenomena such as the spectrum of black-body radiation, the photoelectric effect, and the stability and spectra of atoms.", "requested_types": ["MISC", "SCIENTIST", "THEORY", "ENZYME"], "distractors": ["ENZYME"], "gold": {"MISC": ["black-body radiation"], "SCIENTIST": ["Max Planck", "Albert Einstein", "Niels Bohr"], "THEORY": ["photoelectric effect"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST, THEORY, ENZYME\n\nText:\nMax Planck, Albert Einstein, and Niels Bohr postulated the occurrence of energy in discrete quantities (quanta) in order to explain phenomena such as the spectrum of black-body radiation, the photoelectric effect, and the stability and spectra of atoms.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST, THEORY, ENZYME\n\nText:\nMax Planck, Albert Einstein, and Niels Bohr postulated the occurrence of energy in discrete quantities (quanta) in order to explain phenomena such as the spectrum of black-body radiation, the photoelectric effect, and the stability and spectra of atoms.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_272", "text": "The introduction of the Others featured Tom, a.k.a. Mr. Friendly (M. C. Gainey), and Ethan Rom (William Mapother), all of whom have been shown in both flashbacks and the ongoing story.", "requested_types": ["ACADEMIC_JOURNAL", "PERSON"], "distractors": ["ACADEMIC_JOURNAL"], "gold": {"ACADEMIC_JOURNAL": [], "PERSON": ["Tom", "M. C. Gainey", "Ethan Rom", "William Mapother"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, PERSON\n\nText:\nThe introduction of the Others featured Tom, a.k.a. Mr. Friendly (M. C. Gainey), and Ethan Rom (William Mapother), all of whom have been shown in both flashbacks and the ongoing story.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, PERSON\n\nText:\nThe introduction of the Others featured Tom, a.k.a. Mr. Friendly (M. C. Gainey), and Ethan Rom (William Mapother), all of whom have been shown in both flashbacks and the ongoing story.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_273", "text": "When the Royal Society in London published the groundbreaking work of an Italian lensmaker in their journal Philosophical Transactions of the Royal Society, de Graaf wrote to the editor of the journal, Henry Oldenburg, with a ringing endorsement of van Leeuwenhoek's microscopes which, he claimed, far surpass those which we have hitherto seen.", "requested_types": ["ORGANISATION", "MISC", "COUNTRY", "SCIENTIST", "AWARD", "PROTEIN", "ACADEMIC_JOURNAL", "LOCATION"], "distractors": ["COUNTRY", "AWARD", "PROTEIN"], "gold": {"ORGANISATION": ["Royal Society"], "MISC": ["Italian"], "COUNTRY": [], "SCIENTIST": ["Henry Oldenburg", "van Leeuwenhoek"], "AWARD": [], "PROTEIN": [], "ACADEMIC_JOURNAL": ["Philosophical Transactions of the Royal Society"], "LOCATION": ["London"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, COUNTRY, SCIENTIST, AWARD, PROTEIN, ACADEMIC_JOURNAL, LOCATION\n\nText:\nWhen the Royal Society in London published the groundbreaking work of an Italian lensmaker in their journal Philosophical Transactions of the Royal Society, de Graaf wrote to the editor of the journal, Henry Oldenburg, with a ringing endorsement of van Leeuwenhoek's microscopes which, he claimed, far surpass those which we have hitherto seen.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, COUNTRY, SCIENTIST, AWARD, PROTEIN, ACADEMIC_JOURNAL, LOCATION\n\nText:\nWhen the Royal Society in London published the groundbreaking work of an Italian lensmaker in their journal Philosophical Transactions of the Royal Society, de Graaf wrote to the editor of the journal, Henry Oldenburg, with a ringing endorsement of van Leeuwenhoek's microscopes which, he claimed, far surpass those which we have hitherto seen.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_274", "text": "Under his leadership the neutron was discovered by James Chadwick in 1932 and in the same year the first experiment to split the nucleus in a fully controlled manner was performed by students working under his direction, John Cockcroft and Ernest Walton.", "requested_types": ["THEORY", "ACADEMIC_JOURNAL", "DISCIPLINE", "SCIENTIST"], "distractors": ["DISCIPLINE", "THEORY", "ACADEMIC_JOURNAL"], "gold": {"THEORY": [], "ACADEMIC_JOURNAL": [], "DISCIPLINE": [], "SCIENTIST": ["James Chadwick", "John Cockcroft", "Ernest Walton"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ACADEMIC_JOURNAL, DISCIPLINE, SCIENTIST\n\nText:\nUnder his leadership the neutron was discovered by James Chadwick in 1932 and in the same year the first experiment to split the nucleus in a fully controlled manner was performed by students working under his direction, John Cockcroft and Ernest Walton.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ACADEMIC_JOURNAL, DISCIPLINE, SCIENTIST\n\nText:\nUnder his leadership the neutron was discovered by James Chadwick in 1932 and in the same year the first experiment to split the nucleus in a fully controlled manner was performed by students working under his direction, John Cockcroft and Ernest Walton.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_275", "text": "Whereas the xenon fluorides are well characterized, with the exception of dichloride Xenon dichloride and Xenon tetrachloride, the other halides are not known.", "requested_types": ["AWARD", "CHEMICAL_COMPOUND", "PROTEIN"], "distractors": ["AWARD", "PROTEIN"], "gold": {"AWARD": [], "CHEMICAL_COMPOUND": ["xenon fluorides", "dichloride", "Xenon dichloride", "Xenon tetrachloride"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, CHEMICAL_COMPOUND, PROTEIN\n\nText:\nWhereas the xenon fluorides are well characterized, with the exception of dichloride Xenon dichloride and Xenon tetrachloride, the other halides are not known.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, CHEMICAL_COMPOUND, PROTEIN\n\nText:\nWhereas the xenon fluorides are well characterized, with the exception of dichloride Xenon dichloride and Xenon tetrachloride, the other halides are not known.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_276", "text": "As examples, antibiotic drug s of the tetracycline and quinolone families are chelators of Fe sup2 + / sup, Calcium sup2 + / sup, and Magnesium sup2 + / sup ions.", "requested_types": ["PERSON", "CHEMICAL_COMPOUND", "ORGANISATION", "CHEMICAL_ELEMENT"], "distractors": ["PERSON", "ORGANISATION"], "gold": {"PERSON": [], "CHEMICAL_COMPOUND": ["tetracycline", "quinolone"], "ORGANISATION": [], "CHEMICAL_ELEMENT": ["Fe sup2 + / sup", "Calcium sup2 + / sup", "Magnesium sup2 + / sup"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, CHEMICAL_COMPOUND, ORGANISATION, CHEMICAL_ELEMENT\n\nText:\nAs examples, antibiotic drug s of the tetracycline and quinolone families are chelators of Fe sup2 + / sup, Calcium sup2 + / sup, and Magnesium sup2 + / sup ions.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, CHEMICAL_COMPOUND, ORGANISATION, CHEMICAL_ELEMENT\n\nText:\nAs examples, antibiotic drug s of the tetracycline and quinolone families are chelators of Fe sup2 + / sup, Calcium sup2 + / sup, and Magnesium sup2 + / sup ions.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_277", "text": "Precursors of T cells originate in the bone marrow from which they migrate via bloodstream into thymic cortex, where they encounter stromal cell s including cTECs, which form the microenvironment crucial for proliferation and development of T cells by expression of DLL4 (delta-like notch ligand 4), cytokine s IL-7, TGFβ or stem cell factor and chemokine s CCL25, CXCL12 or CCRL1 etc.", "requested_types": ["CHEMICAL_ELEMENT", "CHEMICAL_COMPOUND", "MISC", "PROTEIN", "AWARD", "LOCATION"], "distractors": ["LOCATION", "CHEMICAL_ELEMENT", "AWARD"], "gold": {"CHEMICAL_ELEMENT": [], "CHEMICAL_COMPOUND": ["Precursors"], "MISC": ["T cells", "bone marrow", "thymic cortex", "stromal cell", "cTECs", "cytokine"], "PROTEIN": ["DLL4", "delta-like notch ligand 4", "IL-7", "TGFβ", "stem cell factor", "chemokine", "CCL25", "CXCL12", "CCRL1"], "AWARD": [], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, CHEMICAL_COMPOUND, MISC, PROTEIN, AWARD, LOCATION\n\nText:\nPrecursors of T cells originate in the bone marrow from which they migrate via bloodstream into thymic cortex, where they encounter stromal cell s including cTECs, which form the microenvironment crucial for proliferation and development of T cells by expression of DLL4 (delta-like notch ligand 4), cytokine s IL-7, TGFβ or stem cell factor and chemokine s CCL25, CXCL12 or CCRL1 etc.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, CHEMICAL_COMPOUND, MISC, PROTEIN, AWARD, LOCATION\n\nText:\nPrecursors of T cells originate in the bone marrow from which they migrate via bloodstream into thymic cortex, where they encounter stromal cell s including cTECs, which form the microenvironment crucial for proliferation and development of T cells by expression of DLL4 (delta-like notch ligand 4), cytokine s IL-7, TGFβ or stem cell factor and chemokine s CCL25, CXCL12 or CCRL1 etc.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_278", "text": "Donna J. Haraway (born September 6, 1944) is an American Professor Emerita in the History of Consciousness Department and Feminist Studies Department at the University of California, Santa Cruz, United States.", "requested_types": ["UNIVERSITY", "COUNTRY", "LOCATION", "AWARD", "PERSON", "ORGANISATION", "MISC", "ACADEMIC_JOURNAL", "SCIENTIST"], "distractors": ["PERSON", "ACADEMIC_JOURNAL"], "gold": {"UNIVERSITY": ["University of California"], "COUNTRY": ["United States"], "LOCATION": ["Santa Cruz"], "AWARD": ["Professor Emerita"], "PERSON": [], "ORGANISATION": ["History of Consciousness Department", "Feminist Studies Department"], "MISC": ["American"], "ACADEMIC_JOURNAL": [], "SCIENTIST": ["Donna J. Haraway"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, COUNTRY, LOCATION, AWARD, PERSON, ORGANISATION, MISC, ACADEMIC_JOURNAL, SCIENTIST\n\nText:\nDonna J. Haraway (born September 6, 1944) is an American Professor Emerita in the History of Consciousness Department and Feminist Studies Department at the University of California, Santa Cruz, United States.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, COUNTRY, LOCATION, AWARD, PERSON, ORGANISATION, MISC, ACADEMIC_JOURNAL, SCIENTIST\n\nText:\nDonna J. Haraway (born September 6, 1944) is an American Professor Emerita in the History of Consciousness Department and Feminist Studies Department at the University of California, Santa Cruz, United States.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_279", "text": "Since then, names have been given to 134 additional satellites: 57 satellites of Jupiter, 43 of Saturn, 22 of Uranus, 12 of Neptune, 5 of Pluto, 1 of Eris, and 2 of Haumea.", "requested_types": ["DISCIPLINE", "ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND", "MISC"], "distractors": ["DISCIPLINE", "MISC", "CHEMICAL_COMPOUND"], "gold": {"DISCIPLINE": [], "ASTRONOMICAL_OBJECT": ["Jupiter", "Saturn", "Uranus", "Neptune", "Pluto", "Eris", "Haumea"], "CHEMICAL_COMPOUND": [], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, MISC\n\nText:\nSince then, names have been given to 134 additional satellites: 57 satellites of Jupiter, 43 of Saturn, 22 of Uranus, 12 of Neptune, 5 of Pluto, 1 of Eris, and 2 of Haumea.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, MISC\n\nText:\nSince then, names have been given to 134 additional satellites: 57 satellites of Jupiter, 43 of Saturn, 22 of Uranus, 12 of Neptune, 5 of Pluto, 1 of Eris, and 2 of Haumea.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_280", "text": "His research group's developments in the solar and plasmonics field have been featured in Scientific American and in research papers in Science, Nature Materials, Nature Photonics and Advanced Materials.", "requested_types": ["ASTRONOMICAL_OBJECT", "ACADEMIC_JOURNAL", "THEORY", "DISCIPLINE", "MISC"], "distractors": ["THEORY", "ASTRONOMICAL_OBJECT"], "gold": {"ASTRONOMICAL_OBJECT": [], "ACADEMIC_JOURNAL": ["Science", "Nature Materials", "Nature Photonics", "Advanced Materials"], "THEORY": [], "DISCIPLINE": ["solar and plasmonics field"], "MISC": ["Scientific American"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ACADEMIC_JOURNAL, THEORY, DISCIPLINE, MISC\n\nText:\nHis research group's developments in the solar and plasmonics field have been featured in Scientific American and in research papers in Science, Nature Materials, Nature Photonics and Advanced Materials.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ACADEMIC_JOURNAL, THEORY, DISCIPLINE, MISC\n\nText:\nHis research group's developments in the solar and plasmonics field have been featured in Scientific American and in research papers in Science, Nature Materials, Nature Photonics and Advanced Materials.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_281", "text": "The settlements included Grytviken, Leith Harbour, King Edward Point, Stromness, Husvik, Prince Olav Harbour, Ocean Harbour and Godthul.", "requested_types": ["LOCATION", "ENZYME"], "distractors": ["ENZYME"], "gold": {"LOCATION": ["Grytviken", "Leith Harbour", "King Edward Point", "Stromness", "Husvik", "Prince Olav Harbour", "Ocean Harbour", "Godthul"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ENZYME\n\nText:\nThe settlements included Grytviken, Leith Harbour, King Edward Point, Stromness, Husvik, Prince Olav Harbour, Ocean Harbour and Godthul.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ENZYME\n\nText:\nThe settlements included Grytviken, Leith Harbour, King Edward Point, Stromness, Husvik, Prince Olav Harbour, Ocean Harbour and Godthul.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_282", "text": "Other work included an improved determination of the rotation period of Mars, the discovery that the Martian polar caps vary seasonally, the discovery of Titania and Oberon (moons of Uranus) and Enceladus and Mimas (moons of Saturn).", "requested_types": ["MISC", "ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND"], "gold": {"MISC": ["Martian"], "ASTRONOMICAL_OBJECT": ["Mars", "Titania", "Oberon", "Uranus", "Enceladus", "Mimas", "Saturn"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND\n\nText:\nOther work included an improved determination of the rotation period of Mars, the discovery that the Martian polar caps vary seasonally, the discovery of Titania and Oberon (moons of Uranus) and Enceladus and Mimas (moons of Saturn).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND\n\nText:\nOther work included an improved determination of the rotation period of Mars, the discovery that the Martian polar caps vary seasonally, the discovery of Titania and Oberon (moons of Uranus) and Enceladus and Mimas (moons of Saturn).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_283", "text": "Specifically, in cells treated with Ezh1 / 2 inhibitors, greater numbers of promyelocytic leukemia (PML) foci, which are involved with repression of gene expression, were seen, along with upregulation of interferon-alpha (IFN-α) and Interleukin.", "requested_types": ["MISC", "CHEMICAL_COMPOUND", "ORGANISATION", "PERSON"], "distractors": ["ORGANISATION", "PERSON", "CHEMICAL_COMPOUND"], "gold": {"MISC": ["Ezh1 / 2 inhibitors", "promyelocytic leukemia", "PML", "interferon-alpha", "IFN-α", "Interleukin"], "CHEMICAL_COMPOUND": [], "ORGANISATION": [], "PERSON": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_COMPOUND, ORGANISATION, PERSON\n\nText:\nSpecifically, in cells treated with Ezh1 / 2 inhibitors, greater numbers of promyelocytic leukemia (PML) foci, which are involved with repression of gene expression, were seen, along with upregulation of interferon-alpha (IFN-α) and Interleukin.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_COMPOUND, ORGANISATION, PERSON\n\nText:\nSpecifically, in cells treated with Ezh1 / 2 inhibitors, greater numbers of promyelocytic leukemia (PML) foci, which are involved with repression of gene expression, were seen, along with upregulation of interferon-alpha (IFN-α) and Interleukin.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_284", "text": "The visitor starts with the sun on the eastern summit of the Hill, and following the trail to the west from the Sun, in the direction of Mills Observatory, he or she will encounter another eight rocks representing the planets Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus and Neptune.", "requested_types": ["LOCATION", "ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"LOCATION": ["Hill", "Mills Observatory"], "ASTRONOMICAL_OBJECT": ["sun", "Sun", "Mercury", "Venus", "Earth", "Mars", "Jupiter", "Saturn", "Uranus", "Neptune"], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT\n\nText:\nThe visitor starts with the sun on the eastern summit of the Hill, and following the trail to the west from the Sun, in the direction of Mills Observatory, he or she will encounter another eight rocks representing the planets Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus and Neptune.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT\n\nText:\nThe visitor starts with the sun on the eastern summit of the Hill, and following the trail to the west from the Sun, in the direction of Mills Observatory, he or she will encounter another eight rocks representing the planets Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus and Neptune.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_285", "text": "Simpson's other great passions were astronomy (he was a member of the British Astronomical Association and - unusually for an amateur - was made a Fellow of the Royal Astronomical Society) and pacifism, specifically addressed in the title of his Tenth String Quartet, For Peace.", "requested_types": ["DISCIPLINE", "AWARD", "CHEMICAL_ELEMENT", "ORGANISATION", "PERSON", "MISC"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"DISCIPLINE": ["astronomy"], "AWARD": ["Fellow of the Royal Astronomical Society"], "CHEMICAL_ELEMENT": [], "ORGANISATION": ["British Astronomical Association"], "PERSON": ["Simpson"], "MISC": ["pacifism", "Tenth String Quartet", "For Peace"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, AWARD, CHEMICAL_ELEMENT, ORGANISATION, PERSON, MISC\n\nText:\nSimpson's other great passions were astronomy (he was a member of the British Astronomical Association and - unusually for an amateur - was made a Fellow of the Royal Astronomical Society) and pacifism, specifically addressed in the title of his Tenth String Quartet, For Peace.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, AWARD, CHEMICAL_ELEMENT, ORGANISATION, PERSON, MISC\n\nText:\nSimpson's other great passions were astronomy (he was a member of the British Astronomical Association and - unusually for an amateur - was made a Fellow of the Royal Astronomical Society) and pacifism, specifically addressed in the title of his Tenth String Quartet, For Peace.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_286", "text": "Palsson serves on the editorial board of several scientific journals including Annals of Biomedical Engineering, Biotechnology and Bioengineering, Metabolic Engineering and Molecular Systems Biology.", "requested_types": ["SCIENTIST", "LOCATION", "ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT", "ACADEMIC_JOURNAL"], "distractors": ["CHEMICAL_ELEMENT", "LOCATION", "ASTRONOMICAL_OBJECT"], "gold": {"SCIENTIST": ["Palsson"], "LOCATION": [], "ASTRONOMICAL_OBJECT": [], "CHEMICAL_ELEMENT": [], "ACADEMIC_JOURNAL": ["Annals of Biomedical Engineering", "Biotechnology and Bioengineering", "Metabolic Engineering", "Molecular Systems Biology"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, LOCATION, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, ACADEMIC_JOURNAL\n\nText:\nPalsson serves on the editorial board of several scientific journals including Annals of Biomedical Engineering, Biotechnology and Bioengineering, Metabolic Engineering and Molecular Systems Biology.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, LOCATION, ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, ACADEMIC_JOURNAL\n\nText:\nPalsson serves on the editorial board of several scientific journals including Annals of Biomedical Engineering, Biotechnology and Bioengineering, Metabolic Engineering and Molecular Systems Biology.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_287", "text": "Of the sympatric species, they all contain Egg lysin that drives gamete isolation, but the allopatric species does not.", "requested_types": ["MISC", "COUNTRY", "PROTEIN"], "distractors": ["COUNTRY"], "gold": {"MISC": ["gamete isolation"], "COUNTRY": [], "PROTEIN": ["Egg lysin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, COUNTRY, PROTEIN\n\nText:\nOf the sympatric species, they all contain Egg lysin that drives gamete isolation, but the allopatric species does not.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, COUNTRY, PROTEIN\n\nText:\nOf the sympatric species, they all contain Egg lysin that drives gamete isolation, but the allopatric species does not.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_288", "text": "As examples, for Greeks, Constantine XI Palaiologos and Kolokotronis; and for Serbs, Miloš Obilić and Tzar Lazar; for Montenegrins, Đurađ I Balšić and Ivan Crnojević; for Albanians, George Kastrioti Skanderbeg; for ethnic Macedonians, Nikola Karev for Bulgarians, Vasil Levski, Georgi Sava Rakovski and Hristo Botev and for Croats, Nikola Šubić Zrinjski.", "requested_types": ["ACADEMIC_JOURNAL", "SCIENTIST", "MISC", "ASTRONOMICAL_OBJECT", "PERSON"], "distractors": ["ACADEMIC_JOURNAL", "SCIENTIST", "ASTRONOMICAL_OBJECT"], "gold": {"ACADEMIC_JOURNAL": [], "SCIENTIST": [], "MISC": ["Greeks", "Serbs", "Montenegrins", "Albanians", "ethnic Macedonians", "Bulgarians", "Croats"], "ASTRONOMICAL_OBJECT": [], "PERSON": ["Constantine XI Palaiologos", "Kolokotronis", "Miloš Obilić", "Tzar Lazar", "Đurađ I Balšić", "Ivan Crnojević", "George Kastrioti Skanderbeg", "Nikola Karev", "Vasil Levski", "Georgi Sava Rakovski", "Hristo Botev", "Nikola Šubić Zrinjski"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, MISC, ASTRONOMICAL_OBJECT, PERSON\n\nText:\nAs examples, for Greeks, Constantine XI Palaiologos and Kolokotronis; and for Serbs, Miloš Obilić and Tzar Lazar; for Montenegrins, Đurađ I Balšić and Ivan Crnojević; for Albanians, George Kastrioti Skanderbeg; for ethnic Macedonians, Nikola Karev for Bulgarians, Vasil Levski, Georgi Sava Rakovski and Hristo Botev and for Croats, Nikola Šubić Zrinjski.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, MISC, ASTRONOMICAL_OBJECT, PERSON\n\nText:\nAs examples, for Greeks, Constantine XI Palaiologos and Kolokotronis; and for Serbs, Miloš Obilić and Tzar Lazar; for Montenegrins, Đurađ I Balšić and Ivan Crnojević; for Albanians, George Kastrioti Skanderbeg; for ethnic Macedonians, Nikola Karev for Bulgarians, Vasil Levski, Georgi Sava Rakovski and Hristo Botev and for Croats, Nikola Šubić Zrinjski.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_289", "text": "Today we know that AHR / CYP1-signaling is involved in an amazing array of genetic networks and subcellular processes critical to lifesub production; crosstalk with hypoxia and Hypoxia-inducible factors (HIF) -signaling pathways; transforming growth factor -signaling pathways, as well as growth suppression; tumor initiation; tumor promotion; transgenerational inheritance and epigenetic effects; chromatin remodeling; histone modification; and aging-related and degenerative disease s.", "requested_types": ["ENZYME", "ACADEMIC_JOURNAL", "MISC"], "distractors": ["ACADEMIC_JOURNAL", "ENZYME"], "gold": {"ENZYME": [], "ACADEMIC_JOURNAL": [], "MISC": ["AHR", "CYP1-signaling", "hypoxia", "Hypoxia-inducible factors", "HIF", "transforming growth factor", "growth suppression", "tumor initiation", "tumor promotion", "transgenerational inheritance", "epigenetic effects", "chromatin remodeling", "histone modification", "aging-related and degenerative disease"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ACADEMIC_JOURNAL, MISC\n\nText:\nToday we know that AHR / CYP1-signaling is involved in an amazing array of genetic networks and subcellular processes critical to lifesub production; crosstalk with hypoxia and Hypoxia-inducible factors (HIF) -signaling pathways; transforming growth factor -signaling pathways, as well as growth suppression; tumor initiation; tumor promotion; transgenerational inheritance and epigenetic effects; chromatin remodeling; histone modification; and aging-related and degenerative disease s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ACADEMIC_JOURNAL, MISC\n\nText:\nToday we know that AHR / CYP1-signaling is involved in an amazing array of genetic networks and subcellular processes critical to lifesub production; crosstalk with hypoxia and Hypoxia-inducible factors (HIF) -signaling pathways; transforming growth factor -signaling pathways, as well as growth suppression; tumor initiation; tumor promotion; transgenerational inheritance and epigenetic effects; chromatin remodeling; histone modification; and aging-related and degenerative disease s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_290", "text": "The concept of détournement has had a popular influence amongst contemporary radicals, and the technique can be seen in action in the present day when looking at the work of Culture Jammer s including the Cacophony Society, Billboard Liberation Front, monochrom, Occupy Movements and Adbusters, whose subvertisements detourn.", "requested_types": ["ORGANISATION", "EVENT", "DISCIPLINE", "MISC"], "distractors": ["DISCIPLINE"], "gold": {"ORGANISATION": ["Cacophony Society", "monochrom", "Adbusters"], "EVENT": ["Billboard Liberation Front", "Occupy Movements"], "DISCIPLINE": [], "MISC": ["détournement", "Culture Jammer"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, EVENT, DISCIPLINE, MISC\n\nText:\nThe concept of détournement has had a popular influence amongst contemporary radicals, and the technique can be seen in action in the present day when looking at the work of Culture Jammer s including the Cacophony Society, Billboard Liberation Front, monochrom, Occupy Movements and Adbusters, whose subvertisements detourn.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, EVENT, DISCIPLINE, MISC\n\nText:\nThe concept of détournement has had a popular influence amongst contemporary radicals, and the technique can be seen in action in the present day when looking at the work of Culture Jammer s including the Cacophony Society, Billboard Liberation Front, monochrom, Occupy Movements and Adbusters, whose subvertisements detourn.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_291", "text": "She played a substantial role in the discovery of germ cell migratory pathways (namely those involving gap junction s, G protein-coupled receptor like Tre-1, and isoprenoids), particularly those concerning migration into the ovaries and testis.", "requested_types": ["COUNTRY", "MISC", "PROTEIN"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "MISC": ["gap junction", "ovaries", "testis"], "PROTEIN": ["G protein-coupled receptor", "Tre-1", "isoprenoids"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, MISC, PROTEIN\n\nText:\nShe played a substantial role in the discovery of germ cell migratory pathways (namely those involving gap junction s, G protein-coupled receptor like Tre-1, and isoprenoids), particularly those concerning migration into the ovaries and testis.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, MISC, PROTEIN\n\nText:\nShe played a substantial role in the discovery of germ cell migratory pathways (namely those involving gap junction s, G protein-coupled receptor like Tre-1, and isoprenoids), particularly those concerning migration into the ovaries and testis.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_292", "text": "All of these low-numbered asteroids have numbers between and were discovered between 1876 and the 1930s, predominantly by astronomers Auguste Charlois, Johann Palisa, Max Wolf and Karl Reinmuth (also see category).", "requested_types": ["PROTEIN", "SCIENTIST", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND", "PROTEIN"], "gold": {"PROTEIN": [], "SCIENTIST": ["Auguste Charlois", "Johann Palisa", "Max Wolf", "Karl Reinmuth"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST, CHEMICAL_COMPOUND\n\nText:\nAll of these low-numbered asteroids have numbers between and were discovered between 1876 and the 1930s, predominantly by astronomers Auguste Charlois, Johann Palisa, Max Wolf and Karl Reinmuth (also see category).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST, CHEMICAL_COMPOUND\n\nText:\nAll of these low-numbered asteroids have numbers between and were discovered between 1876 and the 1930s, predominantly by astronomers Auguste Charlois, Johann Palisa, Max Wolf and Karl Reinmuth (also see category).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_293", "text": "It was developed as a treatment for hepatitis C, acting as a NS5B RNA polymerase inhibitor, but while it showed a good safety profile in clinical trials, it was not sufficiently effective to be used as a stand-alone agent.", "requested_types": ["CHEMICAL_COMPOUND", "PROTEIN", "CHEMICAL_ELEMENT", "MISC", "UNIVERSITY"], "distractors": ["PROTEIN", "UNIVERSITY", "CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_COMPOUND": ["NS5B RNA polymerase inhibitor"], "PROTEIN": [], "CHEMICAL_ELEMENT": [], "MISC": ["hepatitis C"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, PROTEIN, CHEMICAL_ELEMENT, MISC, UNIVERSITY\n\nText:\nIt was developed as a treatment for hepatitis C, acting as a NS5B RNA polymerase inhibitor, but while it showed a good safety profile in clinical trials, it was not sufficiently effective to be used as a stand-alone agent.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, PROTEIN, CHEMICAL_ELEMENT, MISC, UNIVERSITY\n\nText:\nIt was developed as a treatment for hepatitis C, acting as a NS5B RNA polymerase inhibitor, but while it showed a good safety profile in clinical trials, it was not sufficiently effective to be used as a stand-alone agent.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_294", "text": "Galaxies and protostar s usually show differential rotation; examples in the Solar System include the Sun, Jupiter and Saturn.", "requested_types": ["MISC", "COUNTRY", "ASTRONOMICAL_OBJECT"], "distractors": ["COUNTRY"], "gold": {"MISC": ["Solar System"], "COUNTRY": [], "ASTRONOMICAL_OBJECT": ["Galaxies", "protostar", "Sun", "Jupiter", "Saturn"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, COUNTRY, ASTRONOMICAL_OBJECT\n\nText:\nGalaxies and protostar s usually show differential rotation; examples in the Solar System include the Sun, Jupiter and Saturn.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, COUNTRY, ASTRONOMICAL_OBJECT\n\nText:\nGalaxies and protostar s usually show differential rotation; examples in the Solar System include the Sun, Jupiter and Saturn.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_295", "text": "Alsufi was discovered on 24 September 1960, by Dutch astronomer couple Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plates taken by Dutch-American astronomer Tom Gehrels at the U.S. Palomar Observatory, California.", "requested_types": ["COUNTRY", "MISC", "ENZYME", "LOCATION", "ACADEMIC_JOURNAL", "SCIENTIST"], "distractors": ["ACADEMIC_JOURNAL", "ENZYME"], "gold": {"COUNTRY": ["U.S."], "MISC": ["Dutch", "Dutch-American"], "ENZYME": [], "LOCATION": ["Leiden", "Palomar Observatory", "California"], "ACADEMIC_JOURNAL": [], "SCIENTIST": ["Alsufi", "Ingrid van Houten-Groeneveld", "Cornelis van Houten", "Tom Gehrels"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, MISC, ENZYME, LOCATION, ACADEMIC_JOURNAL, SCIENTIST\n\nText:\nAlsufi was discovered on 24 September 1960, by Dutch astronomer couple Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plates taken by Dutch-American astronomer Tom Gehrels at the U.S. Palomar Observatory, California.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, MISC, ENZYME, LOCATION, ACADEMIC_JOURNAL, SCIENTIST\n\nText:\nAlsufi was discovered on 24 September 1960, by Dutch astronomer couple Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plates taken by Dutch-American astronomer Tom Gehrels at the U.S. Palomar Observatory, California.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_296", "text": "Manchester hosted the 2002 Commonwealth Games, which left it a legacy of sporting facilities including the City of Manchester Stadium, Manchester Aquatics Centre and the National Cycling Centre, headquarters of British Cycling.", "requested_types": ["PERSON", "LOCATION", "EVENT", "ORGANISATION", "PROTEIN", "SCIENTIST"], "distractors": ["PROTEIN", "SCIENTIST"], "gold": {"PERSON": ["Manchester"], "LOCATION": ["City of Manchester Stadium", "Manchester Aquatics Centre", "National Cycling Centre"], "EVENT": ["2002 Commonwealth Games"], "ORGANISATION": ["British Cycling"], "PROTEIN": [], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, EVENT, ORGANISATION, PROTEIN, SCIENTIST\n\nText:\nManchester hosted the 2002 Commonwealth Games, which left it a legacy of sporting facilities including the City of Manchester Stadium, Manchester Aquatics Centre and the National Cycling Centre, headquarters of British Cycling.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, EVENT, ORGANISATION, PROTEIN, SCIENTIST\n\nText:\nManchester hosted the 2002 Commonwealth Games, which left it a legacy of sporting facilities including the City of Manchester Stadium, Manchester Aquatics Centre and the National Cycling Centre, headquarters of British Cycling.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_297", "text": "The rings of Uranus are a system of ring s around the planet Uranus, intermediate in complexity between the more extensive set around Saturn and the simpler systems around Jupiter and Neptune.", "requested_types": ["COUNTRY", "ORGANISATION", "ASTRONOMICAL_OBJECT"], "distractors": ["COUNTRY", "ORGANISATION"], "gold": {"COUNTRY": [], "ORGANISATION": [], "ASTRONOMICAL_OBJECT": ["Uranus", "Saturn", "Jupiter", "Neptune"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ORGANISATION, ASTRONOMICAL_OBJECT\n\nText:\nThe rings of Uranus are a system of ring s around the planet Uranus, intermediate in complexity between the more extensive set around Saturn and the simpler systems around Jupiter and Neptune.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ORGANISATION, ASTRONOMICAL_OBJECT\n\nText:\nThe rings of Uranus are a system of ring s around the planet Uranus, intermediate in complexity between the more extensive set around Saturn and the simpler systems around Jupiter and Neptune.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_298", "text": "These orbital variations, which are influenced by gravity of the Moon, Sun, Jupiter, and Saturn, form the basis of the Milankovitch cycle.", "requested_types": ["ASTRONOMICAL_OBJECT", "AWARD", "PROTEIN", "MISC", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND", "AWARD", "PROTEIN"], "gold": {"ASTRONOMICAL_OBJECT": ["Moon", "Sun", "Jupiter", "Saturn"], "AWARD": [], "PROTEIN": [], "MISC": ["Milankovitch cycle"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, AWARD, PROTEIN, MISC, CHEMICAL_COMPOUND\n\nText:\nThese orbital variations, which are influenced by gravity of the Moon, Sun, Jupiter, and Saturn, form the basis of the Milankovitch cycle.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, AWARD, PROTEIN, MISC, CHEMICAL_COMPOUND\n\nText:\nThese orbital variations, which are influenced by gravity of the Moon, Sun, Jupiter, and Saturn, form the basis of the Milankovitch cycle.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_299", "text": "Some exceptions are the Taraxacum officinale (Taraxacum officinale), which since its introduction to the region has proliferated on beaches and shores and near newly opened roads on Baffin Island, Barley (Hordeum vulgare), shepherd's purse (Capsella bursa-pastoris) and the opium poppy (Papaver somniferum).", "requested_types": ["LOCATION", "DISCIPLINE", "MISC", "CHEMICAL_ELEMENT", "PROTEIN", "ENZYME"], "distractors": ["CHEMICAL_ELEMENT", "PROTEIN", "DISCIPLINE"], "gold": {"LOCATION": ["Baffin Island"], "DISCIPLINE": [], "MISC": ["Barley", "Hordeum vulgare", "shepherd's purse", "Capsella bursa-pastoris", "opium poppy", "Papaver somniferum"], "CHEMICAL_ELEMENT": [], "PROTEIN": [], "ENZYME": ["Taraxacum officinale"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, DISCIPLINE, MISC, CHEMICAL_ELEMENT, PROTEIN, ENZYME\n\nText:\nSome exceptions are the Taraxacum officinale (Taraxacum officinale), which since its introduction to the region has proliferated on beaches and shores and near newly opened roads on Baffin Island, Barley (Hordeum vulgare), shepherd's purse (Capsella bursa-pastoris) and the opium poppy (Papaver somniferum).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, DISCIPLINE, MISC, CHEMICAL_ELEMENT, PROTEIN, ENZYME\n\nText:\nSome exceptions are the Taraxacum officinale (Taraxacum officinale), which since its introduction to the region has proliferated on beaches and shores and near newly opened roads on Baffin Island, Barley (Hordeum vulgare), shepherd's purse (Capsella bursa-pastoris) and the opium poppy (Papaver somniferum).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_300", "text": "In 1348, Giovanni Dondi built the first known clock driven mechanism which displays the ecliptical position of Moon, Sun, Mercury, Venus, Mars, Jupiter and Saturn according to the complicated Ptolemaic planetary theories.", "requested_types": ["ENZYME", "EVENT", "THEORY", "ASTRONOMICAL_OBJECT", "SCIENTIST", "PROTEIN"], "distractors": ["ENZYME", "EVENT", "PROTEIN"], "gold": {"ENZYME": [], "EVENT": [], "THEORY": ["Ptolemaic planetary theories"], "ASTRONOMICAL_OBJECT": ["Moon", "Sun", "Mercury", "Venus", "Mars", "Jupiter", "Saturn"], "SCIENTIST": ["Giovanni Dondi"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, EVENT, THEORY, ASTRONOMICAL_OBJECT, SCIENTIST, PROTEIN\n\nText:\nIn 1348, Giovanni Dondi built the first known clock driven mechanism which displays the ecliptical position of Moon, Sun, Mercury, Venus, Mars, Jupiter and Saturn according to the complicated Ptolemaic planetary theories.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, EVENT, THEORY, ASTRONOMICAL_OBJECT, SCIENTIST, PROTEIN\n\nText:\nIn 1348, Giovanni Dondi built the first known clock driven mechanism which displays the ecliptical position of Moon, Sun, Mercury, Venus, Mars, Jupiter and Saturn according to the complicated Ptolemaic planetary theories.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_301", "text": "H3K27me3 is an epigenetic modification to the DNA packaging protein Histone H3.", "requested_types": ["MISC", "AWARD", "UNIVERSITY", "CHEMICAL_COMPOUND", "PROTEIN", "ORGANISATION"], "distractors": ["AWARD", "ORGANISATION", "UNIVERSITY"], "gold": {"MISC": ["epigenetic modification", "DNA packaging protein"], "AWARD": [], "UNIVERSITY": [], "CHEMICAL_COMPOUND": ["H3K27me3"], "PROTEIN": ["Histone H3"], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, AWARD, UNIVERSITY, CHEMICAL_COMPOUND, PROTEIN, ORGANISATION\n\nText:\nH3K27me3 is an epigenetic modification to the DNA packaging protein Histone H3.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, AWARD, UNIVERSITY, CHEMICAL_COMPOUND, PROTEIN, ORGANISATION\n\nText:\nH3K27me3 is an epigenetic modification to the DNA packaging protein Histone H3.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_302", "text": "In 1994, Tooze moved to London to set up a lab studying the biogenesis of Secretion at the Imperial Cancer Research Fund, later the Cancer Research UK London Research Institute (and now part of the Francis Crick Institute).", "requested_types": ["PERSON", "SCIENTIST", "PROTEIN", "LOCATION", "THEORY", "ORGANISATION"], "distractors": ["PROTEIN", "PERSON", "THEORY"], "gold": {"PERSON": [], "SCIENTIST": ["Tooze"], "PROTEIN": [], "LOCATION": ["London"], "THEORY": [], "ORGANISATION": ["Imperial Cancer Research Fund", "Cancer Research UK London Research Institute", "Francis Crick Institute"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, SCIENTIST, PROTEIN, LOCATION, THEORY, ORGANISATION\n\nText:\nIn 1994, Tooze moved to London to set up a lab studying the biogenesis of Secretion at the Imperial Cancer Research Fund, later the Cancer Research UK London Research Institute (and now part of the Francis Crick Institute).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, SCIENTIST, PROTEIN, LOCATION, THEORY, ORGANISATION\n\nText:\nIn 1994, Tooze moved to London to set up a lab studying the biogenesis of Secretion at the Imperial Cancer Research Fund, later the Cancer Research UK London Research Institute (and now part of the Francis Crick Institute).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_303", "text": "He also included perturbations due to the other planets (principally Jupiter and Venus) and also accounted for the more difficult problem of the non-spherical nature of the Earth and Moon.", "requested_types": ["ASTRONOMICAL_OBJECT", "LOCATION"], "distractors": ["LOCATION"], "gold": {"ASTRONOMICAL_OBJECT": ["Jupiter", "Venus", "Earth", "Moon"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, LOCATION\n\nText:\nHe also included perturbations due to the other planets (principally Jupiter and Venus) and also accounted for the more difficult problem of the non-spherical nature of the Earth and Moon.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, LOCATION\n\nText:\nHe also included perturbations due to the other planets (principally Jupiter and Venus) and also accounted for the more difficult problem of the non-spherical nature of the Earth and Moon.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_304", "text": "Among the biggest YEC organizations are Answers in Genesis, Institute for Creation Research, and Creation Ministries International.", "requested_types": ["PERSON", "ORGANISATION"], "distractors": ["PERSON"], "gold": {"PERSON": [], "ORGANISATION": ["Answers in Genesis", "Institute for Creation Research", "Creation Ministries International"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION\n\nText:\nAmong the biggest YEC organizations are Answers in Genesis, Institute for Creation Research, and Creation Ministries International.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION\n\nText:\nAmong the biggest YEC organizations are Answers in Genesis, Institute for Creation Research, and Creation Ministries International.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_305", "text": "The transcription factors STAT protein, NF-κB, p73, and p53 are among the many proteins that PIAS interacts with.", "requested_types": ["ACADEMIC_JOURNAL", "ORGANISATION", "PROTEIN", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND", "ACADEMIC_JOURNAL", "ORGANISATION"], "gold": {"ACADEMIC_JOURNAL": [], "ORGANISATION": [], "PROTEIN": ["STAT protein", "NF-κB", "p73", "p53", "PIAS"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, ORGANISATION, PROTEIN, CHEMICAL_COMPOUND\n\nText:\nThe transcription factors STAT protein, NF-κB, p73, and p53 are among the many proteins that PIAS interacts with.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, ORGANISATION, PROTEIN, CHEMICAL_COMPOUND\n\nText:\nThe transcription factors STAT protein, NF-κB, p73, and p53 are among the many proteins that PIAS interacts with.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_306", "text": "The journal is published by Taylor and Francis on behalf of the American Association of Geographers, of which it is one of four official journals, the others being The Professional Geographer, AAG Review of Books, GeoHumanities, and African Geographical Review.", "requested_types": ["CHEMICAL_COMPOUND", "ACADEMIC_JOURNAL", "THEORY", "AWARD", "ORGANISATION"], "distractors": ["CHEMICAL_COMPOUND", "THEORY", "AWARD"], "gold": {"CHEMICAL_COMPOUND": [], "ACADEMIC_JOURNAL": ["The Professional Geographer", "AAG Review of Books", "GeoHumanities", "African Geographical Review"], "THEORY": [], "AWARD": [], "ORGANISATION": ["Taylor and Francis", "American Association of Geographers"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ACADEMIC_JOURNAL, THEORY, AWARD, ORGANISATION\n\nText:\nThe journal is published by Taylor and Francis on behalf of the American Association of Geographers, of which it is one of four official journals, the others being The Professional Geographer, AAG Review of Books, GeoHumanities, and African Geographical Review.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ACADEMIC_JOURNAL, THEORY, AWARD, ORGANISATION\n\nText:\nThe journal is published by Taylor and Francis on behalf of the American Association of Geographers, of which it is one of four official journals, the others being The Professional Geographer, AAG Review of Books, GeoHumanities, and African Geographical Review.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_307", "text": "The scientists reported that genome editing by CRISPR induced DNA damage response and the cell cycle stopped.", "requested_types": ["MISC", "ORGANISATION", "COUNTRY"], "distractors": ["COUNTRY", "ORGANISATION"], "gold": {"MISC": ["genome editing", "CRISPR", "DNA damage response"], "ORGANISATION": [], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ORGANISATION, COUNTRY\n\nText:\nThe scientists reported that genome editing by CRISPR induced DNA damage response and the cell cycle stopped.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ORGANISATION, COUNTRY\n\nText:\nThe scientists reported that genome editing by CRISPR induced DNA damage response and the cell cycle stopped.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_308", "text": "Jared Diamond describes an Evil Quartet of habitat destruction, overkill, introduced species and secondary extinctions.small which has been adopted by major international conservation organizations such as the US Nature Conservancy, the World Wildlife Fund, Conservation International and BirdLife International.", "requested_types": ["PROTEIN", "MISC", "SCIENTIST", "ORGANISATION", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND", "PROTEIN"], "gold": {"PROTEIN": [], "MISC": ["Evil Quartet", "habitat destruction"], "SCIENTIST": ["Jared Diamond"], "ORGANISATION": ["US Nature Conservancy", "World Wildlife Fund", "Conservation International", "BirdLife International"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, SCIENTIST, ORGANISATION, CHEMICAL_COMPOUND\n\nText:\nJared Diamond describes an Evil Quartet of habitat destruction, overkill, introduced species and secondary extinctions.small which has been adopted by major international conservation organizations such as the US Nature Conservancy, the World Wildlife Fund, Conservation International and BirdLife International.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, SCIENTIST, ORGANISATION, CHEMICAL_COMPOUND\n\nText:\nJared Diamond describes an Evil Quartet of habitat destruction, overkill, introduced species and secondary extinctions.small which has been adopted by major international conservation organizations such as the US Nature Conservancy, the World Wildlife Fund, Conservation International and BirdLife International.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_309", "text": "Typical COC material will have a higher modulus than HDPE and Polypropylene, similar to Polyethylene terephthalate or PC.", "requested_types": ["DISCIPLINE", "CHEMICAL_COMPOUND"], "distractors": ["DISCIPLINE"], "gold": {"DISCIPLINE": [], "CHEMICAL_COMPOUND": ["COC", "HDPE", "Polypropylene", "Polyethylene terephthalate", "PC"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, CHEMICAL_COMPOUND\n\nText:\nTypical COC material will have a higher modulus than HDPE and Polypropylene, similar to Polyethylene terephthalate or PC.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, CHEMICAL_COMPOUND\n\nText:\nTypical COC material will have a higher modulus than HDPE and Polypropylene, similar to Polyethylene terephthalate or PC.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_310", "text": "Red Mars won the BSFA Award in 1992 and Nebula Award for Best Novel in 1993.", "requested_types": ["AWARD", "MISC", "ORGANISATION", "THEORY"], "distractors": ["THEORY", "ORGANISATION"], "gold": {"AWARD": ["BSFA Award", "Nebula Award for Best Novel"], "MISC": ["Red Mars"], "ORGANISATION": [], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, MISC, ORGANISATION, THEORY\n\nText:\nRed Mars won the BSFA Award in 1992 and Nebula Award for Best Novel in 1993.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, MISC, ORGANISATION, THEORY\n\nText:\nRed Mars won the BSFA Award in 1992 and Nebula Award for Best Novel in 1993.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_311", "text": "1 The eight planets are: Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus, and Neptune.", "requested_types": ["ASTRONOMICAL_OBJECT", "ENZYME"], "distractors": ["ENZYME"], "gold": {"ASTRONOMICAL_OBJECT": ["Mercury", "Venus", "Earth", "Mars", "Jupiter", "Saturn", "Uranus", "Neptune"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ENZYME\n\nText:\n1 The eight planets are: Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus, and Neptune.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ENZYME\n\nText:\n1 The eight planets are: Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus, and Neptune.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_312", "text": "Aryl hydrocarbon receptor repressor and F2RL3 are similarly hypomethylated in smokers.", "requested_types": ["ORGANISATION", "MISC", "ASTRONOMICAL_OBJECT"], "distractors": ["ASTRONOMICAL_OBJECT", "ORGANISATION"], "gold": {"ORGANISATION": [], "MISC": ["Aryl hydrocarbon receptor repressor", "F2RL3"], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, ASTRONOMICAL_OBJECT\n\nText:\nAryl hydrocarbon receptor repressor and F2RL3 are similarly hypomethylated in smokers.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, ASTRONOMICAL_OBJECT\n\nText:\nAryl hydrocarbon receptor repressor and F2RL3 are similarly hypomethylated in smokers.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_313", "text": "These include the TRUE Blue Crew, Antipodean Resistance, the Australian Defence League, National Action (Australia), the Q Society, Reclaim Australia and the Lads Society (formerly United Patriots Front).", "requested_types": ["COUNTRY", "ORGANISATION", "ACADEMIC_JOURNAL"], "distractors": ["ACADEMIC_JOURNAL"], "gold": {"COUNTRY": ["Australia"], "ORGANISATION": ["TRUE Blue Crew", "Antipodean Resistance", "Australian Defence League", "National Action", "Q Society", "Reclaim Australia", "Lads Society", "United Patriots Front"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ORGANISATION, ACADEMIC_JOURNAL\n\nText:\nThese include the TRUE Blue Crew, Antipodean Resistance, the Australian Defence League, National Action (Australia), the Q Society, Reclaim Australia and the Lads Society (formerly United Patriots Front).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ORGANISATION, ACADEMIC_JOURNAL\n\nText:\nThese include the TRUE Blue Crew, Antipodean Resistance, the Australian Defence League, National Action (Australia), the Q Society, Reclaim Australia and the Lads Society (formerly United Patriots Front).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_314", "text": "He was named one of the most valued reviewers of 2010 by the Editors of Elsevier and Nuclear Physics Other journals for which Dr. Poenaru peer review ed articles include Physical Review Letters, Physical Review C, Journal of Physics G: Nuclear and Particle Physics and Canadian Journal of Physics.", "requested_types": ["SCIENTIST", "MISC", "ORGANISATION", "ACADEMIC_JOURNAL"], "distractors": ["MISC"], "gold": {"SCIENTIST": ["Poenaru"], "MISC": [], "ORGANISATION": ["Elsevier"], "ACADEMIC_JOURNAL": ["Nuclear Physics", "Physical Review Letters", "Physical Review C", "Journal of Physics G", "Nuclear and Particle Physics", "Canadian Journal of Physics"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, MISC, ORGANISATION, ACADEMIC_JOURNAL\n\nText:\nHe was named one of the most valued reviewers of 2010 by the Editors of Elsevier and Nuclear Physics Other journals for which Dr. Poenaru peer review ed articles include Physical Review Letters, Physical Review C, Journal of Physics G: Nuclear and Particle Physics and Canadian Journal of Physics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, MISC, ORGANISATION, ACADEMIC_JOURNAL\n\nText:\nHe was named one of the most valued reviewers of 2010 by the Editors of Elsevier and Nuclear Physics Other journals for which Dr. Poenaru peer review ed articles include Physical Review Letters, Physical Review C, Journal of Physics G: Nuclear and Particle Physics and Canadian Journal of Physics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_315", "text": "Several TLR immunodeficiencies have been described in which cellular proteins that should transmit the message from the Toll-like receptor to the nucleus are abnormal.", "requested_types": ["PROTEIN", "SCIENTIST"], "distractors": ["SCIENTIST"], "gold": {"PROTEIN": ["TLR", "Toll-like receptor"], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST\n\nText:\nSeveral TLR immunodeficiencies have been described in which cellular proteins that should transmit the message from the Toll-like receptor to the nucleus are abnormal.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST\n\nText:\nSeveral TLR immunodeficiencies have been described in which cellular proteins that should transmit the message from the Toll-like receptor to the nucleus are abnormal.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_316", "text": "52872 Okyrhoe is a centaur orbiting in the outer Solar System between Jupiter and Saturn.", "requested_types": ["UNIVERSITY", "MISC", "THEORY", "ASTRONOMICAL_OBJECT", "ORGANISATION", "ACADEMIC_JOURNAL"], "distractors": ["THEORY", "UNIVERSITY", "ORGANISATION"], "gold": {"UNIVERSITY": [], "MISC": ["centaur", "outer Solar System"], "THEORY": [], "ASTRONOMICAL_OBJECT": ["Jupiter", "Saturn"], "ORGANISATION": [], "ACADEMIC_JOURNAL": ["52872 Okyrhoe"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, MISC, THEORY, ASTRONOMICAL_OBJECT, ORGANISATION, ACADEMIC_JOURNAL\n\nText:\n52872 Okyrhoe is a centaur orbiting in the outer Solar System between Jupiter and Saturn.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, MISC, THEORY, ASTRONOMICAL_OBJECT, ORGANISATION, ACADEMIC_JOURNAL\n\nText:\n52872 Okyrhoe is a centaur orbiting in the outer Solar System between Jupiter and Saturn.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_317", "text": "For example, LHsub2 / sub / LOx bipropellant produces higher Isubsp / sub but lower thrust than RP-1 / LOx due to the exhaust gases having a lower density and higher velocity (Properties of water vs Carbon dioxide and Hsub2 / subO).", "requested_types": ["CHEMICAL_COMPOUND", "DISCIPLINE"], "distractors": ["DISCIPLINE"], "gold": {"CHEMICAL_COMPOUND": ["LHsub2 / sub / LOx", "Isubsp / sub", "RP-1 / LOx", "water", "Carbon dioxide", "Hsub2 / subO"], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, DISCIPLINE\n\nText:\nFor example, LHsub2 / sub / LOx bipropellant produces higher Isubsp / sub but lower thrust than RP-1 / LOx due to the exhaust gases having a lower density and higher velocity (Properties of water vs Carbon dioxide and Hsub2 / subO).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, DISCIPLINE\n\nText:\nFor example, LHsub2 / sub / LOx bipropellant produces higher Isubsp / sub but lower thrust than RP-1 / LOx due to the exhaust gases having a lower density and higher velocity (Properties of water vs Carbon dioxide and Hsub2 / subO).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_318", "text": "The term can also be used to describe the motion of a satellite across its parent planet, for instance one of the Galilean satellites (Io, Europa, Ganymede, Callisto) across Jupiter, as seen from Earth.", "requested_types": ["DISCIPLINE", "ACADEMIC_JOURNAL", "MISC", "UNIVERSITY", "ASTRONOMICAL_OBJECT"], "distractors": ["ACADEMIC_JOURNAL", "DISCIPLINE", "UNIVERSITY"], "gold": {"DISCIPLINE": [], "ACADEMIC_JOURNAL": [], "MISC": ["Galilean satellites"], "UNIVERSITY": [], "ASTRONOMICAL_OBJECT": ["Io", "Europa", "Ganymede", "Callisto", "Jupiter", "Earth"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ACADEMIC_JOURNAL, MISC, UNIVERSITY, ASTRONOMICAL_OBJECT\n\nText:\nThe term can also be used to describe the motion of a satellite across its parent planet, for instance one of the Galilean satellites (Io, Europa, Ganymede, Callisto) across Jupiter, as seen from Earth.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ACADEMIC_JOURNAL, MISC, UNIVERSITY, ASTRONOMICAL_OBJECT\n\nText:\nThe term can also be used to describe the motion of a satellite across its parent planet, for instance one of the Galilean satellites (Io, Europa, Ganymede, Callisto) across Jupiter, as seen from Earth.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_319", "text": "For the Good Times (Ray Price) won Song of the Year in 1970 from the Academy of Country Music, while Sunday Morning Coming Down (Johnny Cash) won the same award from the Academy's rival, the Country Music Association, in the same year.", "requested_types": ["PERSON", "AWARD", "COUNTRY", "LOCATION", "ORGANISATION", "THEORY", "MISC"], "distractors": ["COUNTRY", "THEORY", "LOCATION"], "gold": {"PERSON": ["Ray Price", "Johnny Cash"], "AWARD": ["Song of the Year"], "COUNTRY": [], "LOCATION": [], "ORGANISATION": ["Academy of Country Music", "Country Music Association"], "THEORY": [], "MISC": ["Good Times", "Sunday Morning Coming Down"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, AWARD, COUNTRY, LOCATION, ORGANISATION, THEORY, MISC\n\nText:\nFor the Good Times (Ray Price) won Song of the Year in 1970 from the Academy of Country Music, while Sunday Morning Coming Down (Johnny Cash) won the same award from the Academy's rival, the Country Music Association, in the same year.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, AWARD, COUNTRY, LOCATION, ORGANISATION, THEORY, MISC\n\nText:\nFor the Good Times (Ray Price) won Song of the Year in 1970 from the Academy of Country Music, while Sunday Morning Coming Down (Johnny Cash) won the same award from the Academy's rival, the Country Music Association, in the same year.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_320", "text": "It was discovered during the Palomar-Leiden survey on 24 September 1960, by Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, and Tom Gehrels at Palomar Observatory in California, United States.", "requested_types": ["PERSON", "AWARD", "MISC", "COUNTRY", "LOCATION", "SCIENTIST"], "distractors": ["AWARD", "PERSON"], "gold": {"PERSON": [], "AWARD": [], "MISC": ["Palomar-Leiden survey"], "COUNTRY": ["United States"], "LOCATION": ["Leiden", "Palomar Observatory", "California"], "SCIENTIST": ["Ingrid van Houten-Groeneveld", "Cornelis van Houten", "Tom Gehrels"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, AWARD, MISC, COUNTRY, LOCATION, SCIENTIST\n\nText:\nIt was discovered during the Palomar-Leiden survey on 24 September 1960, by Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, and Tom Gehrels at Palomar Observatory in California, United States.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, AWARD, MISC, COUNTRY, LOCATION, SCIENTIST\n\nText:\nIt was discovered during the Palomar-Leiden survey on 24 September 1960, by Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, and Tom Gehrels at Palomar Observatory in California, United States.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_321", "text": "The British English spelling of this compound is hydrogen sulphide, but this spelling is not recommended by the International Union of Pure and Applied Chemistry (IUPAC) or the Royal Society of Chemistry.", "requested_types": ["CHEMICAL_COMPOUND", "MISC", "ORGANISATION", "EVENT"], "distractors": ["EVENT"], "gold": {"CHEMICAL_COMPOUND": ["hydrogen sulphide"], "MISC": ["British English"], "ORGANISATION": ["International Union of Pure and Applied Chemistry", "IUPAC", "Royal Society of Chemistry"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, MISC, ORGANISATION, EVENT\n\nText:\nThe British English spelling of this compound is hydrogen sulphide, but this spelling is not recommended by the International Union of Pure and Applied Chemistry (IUPAC) or the Royal Society of Chemistry.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, MISC, ORGANISATION, EVENT\n\nText:\nThe British English spelling of this compound is hydrogen sulphide, but this spelling is not recommended by the International Union of Pure and Applied Chemistry (IUPAC) or the Royal Society of Chemistry.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_322", "text": "They include the N-terminal scaffold attachment factor -A / B, acinus and PIAS (SAP) domain, the Proline - Isoleucine - Asparagine -Ile- Threonine (PINIT) motif, the RING finger domain - Zinc finger -like zinc -binding domain (RLD), the highly acidic domain (AD), the SUMO-interacting motif (SIM), and the serine / threonine -rich C-terminal region (S / T).", "requested_types": ["PERSON", "ORGANISATION", "CHEMICAL_COMPOUND", "PROTEIN", "MISC", "THEORY"], "distractors": ["THEORY", "ORGANISATION", "PERSON"], "gold": {"PERSON": [], "ORGANISATION": [], "CHEMICAL_COMPOUND": ["Proline", "Isoleucine", "Asparagine", "Threonine", "PINIT"], "PROTEIN": ["N-terminal scaffold attachment factor -A / B", "PIAS (SAP)", "RING finger domain", "Zinc finger -like zinc -binding domain", "RLD", "SUMO-interacting motif", "SIM", "serine / threonine -rich C-terminal region"], "MISC": ["acinus"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION, CHEMICAL_COMPOUND, PROTEIN, MISC, THEORY\n\nText:\nThey include the N-terminal scaffold attachment factor -A / B, acinus and PIAS (SAP) domain, the Proline - Isoleucine - Asparagine -Ile- Threonine (PINIT) motif, the RING finger domain - Zinc finger -like zinc -binding domain (RLD), the highly acidic domain (AD), the SUMO-interacting motif (SIM), and the serine / threonine -rich C-terminal region (S / T).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION, CHEMICAL_COMPOUND, PROTEIN, MISC, THEORY\n\nText:\nThey include the N-terminal scaffold attachment factor -A / B, acinus and PIAS (SAP) domain, the Proline - Isoleucine - Asparagine -Ile- Threonine (PINIT) motif, the RING finger domain - Zinc finger -like zinc -binding domain (RLD), the highly acidic domain (AD), the SUMO-interacting motif (SIM), and the serine / threonine -rich C-terminal region (S / T).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_323", "text": "The use of 1,4-butanediol as an intelligent cosubstrate has also been validated in non-aqueous media using a commercial Alcohol dehydrogenase .Kara S, Spickermann D, Weckbecker A, Leggewie C, Arends IWCE, Hollmann F (2014) Bioreductions Catalyzed by an Alcohol Dehydrogenase in Non-aqueous Media.", "requested_types": ["SCIENTIST", "CHEMICAL_ELEMENT", "UNIVERSITY", "ASTRONOMICAL_OBJECT", "MISC", "CHEMICAL_COMPOUND", "ENZYME"], "distractors": ["ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT", "UNIVERSITY"], "gold": {"SCIENTIST": [".Kara S", "Spickermann D", "Weckbecker A", "Leggewie C", "Arends IWCE", "Hollmann F"], "CHEMICAL_ELEMENT": [], "UNIVERSITY": [], "ASTRONOMICAL_OBJECT": [], "MISC": ["intelligent cosubstrate", "non-aqueous media", "Bioreductions Catalyzed", "Non-aqueous Media"], "CHEMICAL_COMPOUND": ["1,4-butanediol"], "ENZYME": ["Alcohol dehydrogenase", "Alcohol Dehydrogenase"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_ELEMENT, UNIVERSITY, ASTRONOMICAL_OBJECT, MISC, CHEMICAL_COMPOUND, ENZYME\n\nText:\nThe use of 1,4-butanediol as an intelligent cosubstrate has also been validated in non-aqueous media using a commercial Alcohol dehydrogenase .Kara S, Spickermann D, Weckbecker A, Leggewie C, Arends IWCE, Hollmann F (2014) Bioreductions Catalyzed by an Alcohol Dehydrogenase in Non-aqueous Media.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_ELEMENT, UNIVERSITY, ASTRONOMICAL_OBJECT, MISC, CHEMICAL_COMPOUND, ENZYME\n\nText:\nThe use of 1,4-butanediol as an intelligent cosubstrate has also been validated in non-aqueous media using a commercial Alcohol dehydrogenase .Kara S, Spickermann D, Weckbecker A, Leggewie C, Arends IWCE, Hollmann F (2014) Bioreductions Catalyzed by an Alcohol Dehydrogenase in Non-aqueous Media.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_324", "text": "Also dealt with was Miranda Bailey (Chandra Wilson) attempting to have Stevens rejoin the program, Mark Sloan (Dane) arriving from New York City in order to resume his romantic relationship with Montgomery, and Callie Torres (Sara Ramirez) being evacuated from the hospital.", "requested_types": ["COUNTRY", "LOCATION", "CHEMICAL_ELEMENT", "PERSON", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_ELEMENT", "COUNTRY", "CHEMICAL_COMPOUND"], "gold": {"COUNTRY": [], "LOCATION": ["New York City"], "CHEMICAL_ELEMENT": [], "PERSON": ["Miranda Bailey", "Chandra Wilson", "Stevens", "Mark Sloan", "Dane", "Montgomery", "Callie Torres", "Sara Ramirez"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, LOCATION, CHEMICAL_ELEMENT, PERSON, CHEMICAL_COMPOUND\n\nText:\nAlso dealt with was Miranda Bailey (Chandra Wilson) attempting to have Stevens rejoin the program, Mark Sloan (Dane) arriving from New York City in order to resume his romantic relationship with Montgomery, and Callie Torres (Sara Ramirez) being evacuated from the hospital.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, LOCATION, CHEMICAL_ELEMENT, PERSON, CHEMICAL_COMPOUND\n\nText:\nAlso dealt with was Miranda Bailey (Chandra Wilson) attempting to have Stevens rejoin the program, Mark Sloan (Dane) arriving from New York City in order to resume his romantic relationship with Montgomery, and Callie Torres (Sara Ramirez) being evacuated from the hospital.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_325", "text": "Members of this second group are ZONAB, Ral-A, Raf-1, Protein kinase C, symplekin, cingulin and some more.", "requested_types": ["PROTEIN", "ENZYME", "COUNTRY", "MISC"], "distractors": ["COUNTRY"], "gold": {"PROTEIN": ["Ral-A", "Raf-1", "symplekin", "cingulin"], "ENZYME": ["Protein kinase C"], "COUNTRY": [], "MISC": ["ZONAB"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ENZYME, COUNTRY, MISC\n\nText:\nMembers of this second group are ZONAB, Ral-A, Raf-1, Protein kinase C, symplekin, cingulin and some more.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ENZYME, COUNTRY, MISC\n\nText:\nMembers of this second group are ZONAB, Ral-A, Raf-1, Protein kinase C, symplekin, cingulin and some more.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_326", "text": "Amar Nath Bhaduri (1935-2003) was an Indian molecular enzymologist and chemical biologist, known for his studies on UDP-glucose 4-epimerase, a homodimeric epimerase found in cells and his work on Leishmania donovani, the protozoal pathogen for Kala-azar.", "requested_types": ["MISC", "DISCIPLINE", "SCIENTIST", "COUNTRY", "LOCATION", "ENZYME"], "distractors": ["LOCATION", "DISCIPLINE"], "gold": {"MISC": ["molecular enzymologist", "chemical biologist", "Leishmania donovani", "protozoal pathogen", "Kala-azar"], "DISCIPLINE": [], "SCIENTIST": ["Amar Nath Bhaduri"], "COUNTRY": ["Indian"], "LOCATION": [], "ENZYME": ["UDP-glucose 4-epimerase", "homodimeric epimerase"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, DISCIPLINE, SCIENTIST, COUNTRY, LOCATION, ENZYME\n\nText:\nAmar Nath Bhaduri (1935-2003) was an Indian molecular enzymologist and chemical biologist, known for his studies on UDP-glucose 4-epimerase, a homodimeric epimerase found in cells and his work on Leishmania donovani, the protozoal pathogen for Kala-azar.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, DISCIPLINE, SCIENTIST, COUNTRY, LOCATION, ENZYME\n\nText:\nAmar Nath Bhaduri (1935-2003) was an Indian molecular enzymologist and chemical biologist, known for his studies on UDP-glucose 4-epimerase, a homodimeric epimerase found in cells and his work on Leishmania donovani, the protozoal pathogen for Kala-azar.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_327", "text": "Sakharov was discovered on 24 September 1960, by the Dutch astronomers Ingrid van Houten-Groeneveld and Cornelis van Houten, on photographic plates taken by Dutch-American astronomer Tom Gehrels at the U.S. Palomar Observatory in California.", "requested_types": ["SCIENTIST", "ACADEMIC_JOURNAL", "LOCATION", "MISC", "PERSON", "ASTRONOMICAL_OBJECT", "AWARD", "CHEMICAL_COMPOUND"], "distractors": ["AWARD", "CHEMICAL_COMPOUND", "ACADEMIC_JOURNAL"], "gold": {"SCIENTIST": ["Ingrid van Houten-Groeneveld", "Cornelis van Houten"], "ACADEMIC_JOURNAL": [], "LOCATION": ["U.S. Palomar Observatory", "California"], "MISC": ["Dutch", "Dutch-American"], "PERSON": ["Tom Gehrels"], "ASTRONOMICAL_OBJECT": ["Sakharov"], "AWARD": [], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ACADEMIC_JOURNAL, LOCATION, MISC, PERSON, ASTRONOMICAL_OBJECT, AWARD, CHEMICAL_COMPOUND\n\nText:\nSakharov was discovered on 24 September 1960, by the Dutch astronomers Ingrid van Houten-Groeneveld and Cornelis van Houten, on photographic plates taken by Dutch-American astronomer Tom Gehrels at the U.S. Palomar Observatory in California.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ACADEMIC_JOURNAL, LOCATION, MISC, PERSON, ASTRONOMICAL_OBJECT, AWARD, CHEMICAL_COMPOUND\n\nText:\nSakharov was discovered on 24 September 1960, by the Dutch astronomers Ingrid van Houten-Groeneveld and Cornelis van Houten, on photographic plates taken by Dutch-American astronomer Tom Gehrels at the U.S. Palomar Observatory in California.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_328", "text": "The traditional emblem for the House of Lancaster is a red rose, the Red Rose of Lancaster, similar to that of the House of York, which is a white rose.", "requested_types": ["ORGANISATION", "MISC", "EVENT"], "distractors": ["EVENT"], "gold": {"ORGANISATION": ["House of Lancaster"], "MISC": ["Red Rose of Lancaster", "House of York"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, EVENT\n\nText:\nThe traditional emblem for the House of Lancaster is a red rose, the Red Rose of Lancaster, similar to that of the House of York, which is a white rose.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, EVENT\n\nText:\nThe traditional emblem for the House of Lancaster is a red rose, the Red Rose of Lancaster, similar to that of the House of York, which is a white rose.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_329", "text": "; the most exclusive sports clubs: CCI, The Willingdon Sports Club as well as Bombay Gymkhana and the most expensive hospitals: Breach Candy Hospital, Bombay Hospital, Jaslok Hospital and Hurkisondas Hospital; in the nation.", "requested_types": ["LOCATION", "ACADEMIC_JOURNAL", "ORGANISATION"], "distractors": ["ACADEMIC_JOURNAL"], "gold": {"LOCATION": ["Breach Candy Hospital", "Bombay Hospital", "Jaslok Hospital"], "ACADEMIC_JOURNAL": [], "ORGANISATION": ["CCI", "The Willingdon Sports Club", "Bombay Gymkhana", "Hurkisondas Hospital"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ACADEMIC_JOURNAL, ORGANISATION\n\nText:\n; the most exclusive sports clubs: CCI, The Willingdon Sports Club as well as Bombay Gymkhana and the most expensive hospitals: Breach Candy Hospital, Bombay Hospital, Jaslok Hospital and Hurkisondas Hospital; in the nation.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ACADEMIC_JOURNAL, ORGANISATION\n\nText:\n; the most exclusive sports clubs: CCI, The Willingdon Sports Club as well as Bombay Gymkhana and the most expensive hospitals: Breach Candy Hospital, Bombay Hospital, Jaslok Hospital and Hurkisondas Hospital; in the nation.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_330", "text": "The NLR proteins are not linear: the NB-ARC domain is sandwiched in between the Leucine-rich repeat and Toll-Interleukin receptor / CC domains.", "requested_types": ["MISC", "PROTEIN", "ACADEMIC_JOURNAL", "ENZYME", "DISCIPLINE"], "distractors": ["ENZYME", "ACADEMIC_JOURNAL", "DISCIPLINE"], "gold": {"MISC": ["NLR proteins", "CC domains"], "PROTEIN": ["NB-ARC domain", "Leucine-rich repeat", "Toll-Interleukin receptor"], "ACADEMIC_JOURNAL": [], "ENZYME": [], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, ACADEMIC_JOURNAL, ENZYME, DISCIPLINE\n\nText:\nThe NLR proteins are not linear: the NB-ARC domain is sandwiched in between the Leucine-rich repeat and Toll-Interleukin receptor / CC domains.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, ACADEMIC_JOURNAL, ENZYME, DISCIPLINE\n\nText:\nThe NLR proteins are not linear: the NB-ARC domain is sandwiched in between the Leucine-rich repeat and Toll-Interleukin receptor / CC domains.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_331", "text": "In season three, two actors were promoted from recurring to starring roles: Henry Ian Cusick as former Scottish soldier Desmond Hume; and Michael Emerson as the manipulative leader of the Others, Ben Linus.", "requested_types": ["ORGANISATION", "PERSON", "ASTRONOMICAL_OBJECT", "COUNTRY", "MISC"], "distractors": ["ASTRONOMICAL_OBJECT", "COUNTRY", "ORGANISATION"], "gold": {"ORGANISATION": [], "PERSON": ["Henry Ian Cusick", "Desmond Hume", "Michael Emerson", "Ben Linus"], "ASTRONOMICAL_OBJECT": [], "COUNTRY": [], "MISC": ["Scottish"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PERSON, ASTRONOMICAL_OBJECT, COUNTRY, MISC\n\nText:\nIn season three, two actors were promoted from recurring to starring roles: Henry Ian Cusick as former Scottish soldier Desmond Hume; and Michael Emerson as the manipulative leader of the Others, Ben Linus.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PERSON, ASTRONOMICAL_OBJECT, COUNTRY, MISC\n\nText:\nIn season three, two actors were promoted from recurring to starring roles: Henry Ian Cusick as former Scottish soldier Desmond Hume; and Michael Emerson as the manipulative leader of the Others, Ben Linus.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_332", "text": "From left to right, they are: Saturn with sickle and club for Saturday, Jupiter with thunderbolts for Thursday, Mars with sword and shield for Tuesday, Venus with Cupid for Friday and Mercury with staff and bag for Wednesday.", "requested_types": ["ASTRONOMICAL_OBJECT", "AWARD", "EVENT"], "distractors": ["AWARD", "EVENT"], "gold": {"ASTRONOMICAL_OBJECT": ["Saturn", "Jupiter", "Mars", "Venus", "Mercury"], "AWARD": [], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, AWARD, EVENT\n\nText:\nFrom left to right, they are: Saturn with sickle and club for Saturday, Jupiter with thunderbolts for Thursday, Mars with sword and shield for Tuesday, Venus with Cupid for Friday and Mercury with staff and bag for Wednesday.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, AWARD, EVENT\n\nText:\nFrom left to right, they are: Saturn with sickle and club for Saturday, Jupiter with thunderbolts for Thursday, Mars with sword and shield for Tuesday, Venus with Cupid for Friday and Mercury with staff and bag for Wednesday.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_333", "text": "Saha also invented an instrument to measure the weight and pressure of solar rays and helped to build several scientific institutions, such as the Physics Department in University of Allahabad and the Saha Institute of Nuclear Physics in Calcutta.", "requested_types": ["UNIVERSITY", "LOCATION", "ASTRONOMICAL_OBJECT", "ORGANISATION", "PERSON"], "distractors": ["ASTRONOMICAL_OBJECT"], "gold": {"UNIVERSITY": ["University of Allahabad"], "LOCATION": ["Calcutta"], "ASTRONOMICAL_OBJECT": [], "ORGANISATION": ["Physics Department", "Saha Institute of Nuclear Physics"], "PERSON": ["Saha"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, LOCATION, ASTRONOMICAL_OBJECT, ORGANISATION, PERSON\n\nText:\nSaha also invented an instrument to measure the weight and pressure of solar rays and helped to build several scientific institutions, such as the Physics Department in University of Allahabad and the Saha Institute of Nuclear Physics in Calcutta.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, LOCATION, ASTRONOMICAL_OBJECT, ORGANISATION, PERSON\n\nText:\nSaha also invented an instrument to measure the weight and pressure of solar rays and helped to build several scientific institutions, such as the Physics Department in University of Allahabad and the Saha Institute of Nuclear Physics in Calcutta.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_334", "text": "Insulin signaling activates the adenosine triphosphate (ATP) -sensitive potassium (KATP) channels in the arcuate nucleus, decreases AgRP release, and through the vagus nerve, leads to decreased glucose production by the liver by decreasing gluconeogenic enzymes: Phosphoenolpyruvate carboxykinase, Glucose 6-phosphatase.", "requested_types": ["UNIVERSITY", "ENZYME", "ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND", "MISC", "CHEMICAL_ELEMENT", "PROTEIN", "AWARD"], "distractors": ["ASTRONOMICAL_OBJECT", "UNIVERSITY", "AWARD"], "gold": {"UNIVERSITY": [], "ENZYME": ["Phosphoenolpyruvate carboxykinase", "Glucose 6-phosphatase"], "ASTRONOMICAL_OBJECT": [], "CHEMICAL_COMPOUND": ["adenosine triphosphate", "ATP", "AgRP", "glucose"], "MISC": ["arcuate nucleus", "liver", "gluconeogenic enzymes"], "CHEMICAL_ELEMENT": ["potassium", "KATP"], "PROTEIN": ["Insulin"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ENZYME, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, MISC, CHEMICAL_ELEMENT, PROTEIN, AWARD\n\nText:\nInsulin signaling activates the adenosine triphosphate (ATP) -sensitive potassium (KATP) channels in the arcuate nucleus, decreases AgRP release, and through the vagus nerve, leads to decreased glucose production by the liver by decreasing gluconeogenic enzymes: Phosphoenolpyruvate carboxykinase, Glucose 6-phosphatase.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ENZYME, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, MISC, CHEMICAL_ELEMENT, PROTEIN, AWARD\n\nText:\nInsulin signaling activates the adenosine triphosphate (ATP) -sensitive potassium (KATP) channels in the arcuate nucleus, decreases AgRP release, and through the vagus nerve, leads to decreased glucose production by the liver by decreasing gluconeogenic enzymes: Phosphoenolpyruvate carboxykinase, Glucose 6-phosphatase.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_335", "text": "A second meeting was held soon thereafter and included Klaus Clusius, Robert Döpel, Werner Heisenberg, and Carl Friedrich von Weizsäcker.", "requested_types": ["SCIENTIST", "THEORY"], "distractors": ["THEORY"], "gold": {"SCIENTIST": ["Klaus Clusius", "Robert Döpel", "Werner Heisenberg", "Carl Friedrich von Weizsäcker"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, THEORY\n\nText:\nA second meeting was held soon thereafter and included Klaus Clusius, Robert Döpel, Werner Heisenberg, and Carl Friedrich von Weizsäcker.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, THEORY\n\nText:\nA second meeting was held soon thereafter and included Klaus Clusius, Robert Döpel, Werner Heisenberg, and Carl Friedrich von Weizsäcker.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_336", "text": "The conclusion of the scientists was that further effort was needed in to improve the precision and efficiency of CRISPR gene editing.", "requested_types": ["MISC", "SCIENTIST"], "distractors": ["SCIENTIST"], "gold": {"MISC": ["CRISPR gene editing"], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST\n\nText:\nThe conclusion of the scientists was that further effort was needed in to improve the precision and efficiency of CRISPR gene editing.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, SCIENTIST\n\nText:\nThe conclusion of the scientists was that further effort was needed in to improve the precision and efficiency of CRISPR gene editing.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_337", "text": "In the second step react Phosphoribosylamine, glycine and ATP to create Glycineamide ribonucleotide, ADP, and pyrophosphate - catalyzed by phosphoribosylamine - glycine ligase (GAR synthetase).", "requested_types": ["PERSON", "LOCATION", "ENZYME", "ORGANISATION", "CHEMICAL_COMPOUND"], "distractors": ["PERSON", "ORGANISATION", "LOCATION"], "gold": {"PERSON": [], "LOCATION": [], "ENZYME": ["phosphoribosylamine - glycine ligase", "GAR synthetase"], "ORGANISATION": [], "CHEMICAL_COMPOUND": ["Phosphoribosylamine", "glycine", "ATP", "Glycineamide ribonucleotide", "ADP", "pyrophosphate"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, ENZYME, ORGANISATION, CHEMICAL_COMPOUND\n\nText:\nIn the second step react Phosphoribosylamine, glycine and ATP to create Glycineamide ribonucleotide, ADP, and pyrophosphate - catalyzed by phosphoribosylamine - glycine ligase (GAR synthetase).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, LOCATION, ENZYME, ORGANISATION, CHEMICAL_COMPOUND\n\nText:\nIn the second step react Phosphoribosylamine, glycine and ATP to create Glycineamide ribonucleotide, ADP, and pyrophosphate - catalyzed by phosphoribosylamine - glycine ligase (GAR synthetase).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_338", "text": "Schweickart is a fellow of the American Astronautical Society and the International Academy of Astronautics, and an associate fellow of the American Institute of Aeronautics and Astronautics.", "requested_types": ["CHEMICAL_ELEMENT", "AWARD", "ORGANISATION", "SCIENTIST"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_ELEMENT": [], "AWARD": ["fellow of the American Astronautical Society", "fellow of the American Institute of Aeronautics and Astronautics"], "ORGANISATION": ["International Academy of Astronautics"], "SCIENTIST": ["Schweickart"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, AWARD, ORGANISATION, SCIENTIST\n\nText:\nSchweickart is a fellow of the American Astronautical Society and the International Academy of Astronautics, and an associate fellow of the American Institute of Aeronautics and Astronautics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, AWARD, ORGANISATION, SCIENTIST\n\nText:\nSchweickart is a fellow of the American Astronautical Society and the International Academy of Astronautics, and an associate fellow of the American Institute of Aeronautics and Astronautics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_339", "text": "The scientists used tripronuclear (3PN) zygotes, zygotes fertilized by two sperm and therefore non-viable, to investigate CRISPR -mediated gene editing in human cells, something that had never been attempted before.", "requested_types": ["PROTEIN", "MISC"], "distractors": ["PROTEIN"], "gold": {"PROTEIN": [], "MISC": ["tripronuclear (3PN) zygotes", "CRISPR -mediated gene editing"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC\n\nText:\nThe scientists used tripronuclear (3PN) zygotes, zygotes fertilized by two sperm and therefore non-viable, to investigate CRISPR -mediated gene editing in human cells, something that had never been attempted before.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC\n\nText:\nThe scientists used tripronuclear (3PN) zygotes, zygotes fertilized by two sperm and therefore non-viable, to investigate CRISPR -mediated gene editing in human cells, something that had never been attempted before.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_340", "text": "Now being unemployed, Izzie Stevens (Katherine Heigl) tries to move on after the death of her fiancé, Denny Duquette (Jeffrey Dean Morgan).", "requested_types": ["ORGANISATION", "PERSON"], "distractors": ["ORGANISATION"], "gold": {"ORGANISATION": [], "PERSON": ["Izzie Stevens", "Katherine Heigl", "Denny Duquette", "Jeffrey Dean Morgan"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PERSON\n\nText:\nNow being unemployed, Izzie Stevens (Katherine Heigl) tries to move on after the death of her fiancé, Denny Duquette (Jeffrey Dean Morgan).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PERSON\n\nText:\nNow being unemployed, Izzie Stevens (Katherine Heigl) tries to move on after the death of her fiancé, Denny Duquette (Jeffrey Dean Morgan).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_341", "text": "Soon he established closer contact to Wilhelm Wien, Arnold Sommerfeld, Johannes Stark, and Albert Einstein.", "requested_types": ["LOCATION", "CHEMICAL_ELEMENT", "ORGANISATION", "SCIENTIST"], "distractors": ["LOCATION", "CHEMICAL_ELEMENT", "ORGANISATION"], "gold": {"LOCATION": [], "CHEMICAL_ELEMENT": [], "ORGANISATION": [], "SCIENTIST": ["Wilhelm Wien", "Arnold Sommerfeld", "Johannes Stark", "Albert Einstein"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, CHEMICAL_ELEMENT, ORGANISATION, SCIENTIST\n\nText:\nSoon he established closer contact to Wilhelm Wien, Arnold Sommerfeld, Johannes Stark, and Albert Einstein.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, CHEMICAL_ELEMENT, ORGANISATION, SCIENTIST\n\nText:\nSoon he established closer contact to Wilhelm Wien, Arnold Sommerfeld, Johannes Stark, and Albert Einstein.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_342", "text": "He studied astronomy and physics at the University of Dorpat, and furthered his education of physics in Paris in the laboratories of Antoine César Becquerel (1788-1878) and Henri Victor Regnault (1810-1878), and afterwards at Berlin in the laboratories of Heinrich Gustav Magnus (1802-1870), Johann Christian Poggendorff (1796-1877) and Heinrich Wilhelm Dove (1803-1879).", "requested_types": ["UNIVERSITY", "PERSON", "LOCATION", "THEORY", "ORGANISATION", "SCIENTIST", "DISCIPLINE"], "distractors": ["ORGANISATION", "THEORY", "PERSON"], "gold": {"UNIVERSITY": ["University of Dorpat"], "PERSON": [], "LOCATION": ["Paris"], "THEORY": [], "ORGANISATION": [], "SCIENTIST": ["Antoine César Becquerel", "Henri Victor Regnault", "Heinrich Gustav Magnus", "Johann Christian Poggendorff", "Heinrich Wilhelm Dove"], "DISCIPLINE": ["astronomy", "physics"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, PERSON, LOCATION, THEORY, ORGANISATION, SCIENTIST, DISCIPLINE\n\nText:\nHe studied astronomy and physics at the University of Dorpat, and furthered his education of physics in Paris in the laboratories of Antoine César Becquerel (1788-1878) and Henri Victor Regnault (1810-1878), and afterwards at Berlin in the laboratories of Heinrich Gustav Magnus (1802-1870), Johann Christian Poggendorff (1796-1877) and Heinrich Wilhelm Dove (1803-1879).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, PERSON, LOCATION, THEORY, ORGANISATION, SCIENTIST, DISCIPLINE\n\nText:\nHe studied astronomy and physics at the University of Dorpat, and furthered his education of physics in Paris in the laboratories of Antoine César Becquerel (1788-1878) and Henri Victor Regnault (1810-1878), and afterwards at Berlin in the laboratories of Heinrich Gustav Magnus (1802-1870), Johann Christian Poggendorff (1796-1877) and Heinrich Wilhelm Dove (1803-1879).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_343", "text": "Declaring an arsenal of 39,967 tons of chemical weapons in 1997, by far the largest arsenal, consisting of blister agents: Lewisite, Sulfur mustard, Lewisite-mustard mix, and nerve agents: Sarin, Soman, and VX.", "requested_types": ["CHEMICAL_COMPOUND", "THEORY"], "distractors": ["THEORY"], "gold": {"CHEMICAL_COMPOUND": ["blister agents", "Lewisite", "Sulfur mustard", "Lewisite-mustard mix", "nerve agents", "Sarin", "Soman", "VX"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, THEORY\n\nText:\nDeclaring an arsenal of 39,967 tons of chemical weapons in 1997, by far the largest arsenal, consisting of blister agents: Lewisite, Sulfur mustard, Lewisite-mustard mix, and nerve agents: Sarin, Soman, and VX.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, THEORY\n\nText:\nDeclaring an arsenal of 39,967 tons of chemical weapons in 1997, by far the largest arsenal, consisting of blister agents: Lewisite, Sulfur mustard, Lewisite-mustard mix, and nerve agents: Sarin, Soman, and VX.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_344", "text": "Dyson presented his observations of the solar eclipse of May 29, 1919 to a joint meeting of the Royal Society and Royal Astronomical Society on 6 November 1919.", "requested_types": ["SCIENTIST", "EVENT", "ORGANISATION", "UNIVERSITY"], "distractors": ["UNIVERSITY"], "gold": {"SCIENTIST": ["Dyson"], "EVENT": ["solar eclipse of May 29, 1919"], "ORGANISATION": ["Royal Society", "Royal Astronomical Society"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, EVENT, ORGANISATION, UNIVERSITY\n\nText:\nDyson presented his observations of the solar eclipse of May 29, 1919 to a joint meeting of the Royal Society and Royal Astronomical Society on 6 November 1919.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, EVENT, ORGANISATION, UNIVERSITY\n\nText:\nDyson presented his observations of the solar eclipse of May 29, 1919 to a joint meeting of the Royal Society and Royal Astronomical Society on 6 November 1919.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_345", "text": "The discovery of the asteroid 3 Juno by Karl Ludwig Harding and 4 Vesta by Olbers, buttressed his hypothesis.", "requested_types": ["ACADEMIC_JOURNAL", "SCIENTIST", "THEORY", "ASTRONOMICAL_OBJECT"], "distractors": ["ACADEMIC_JOURNAL", "THEORY"], "gold": {"ACADEMIC_JOURNAL": [], "SCIENTIST": ["Karl Ludwig Harding", "Olbers"], "THEORY": [], "ASTRONOMICAL_OBJECT": ["3 Juno", "4 Vesta"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, THEORY, ASTRONOMICAL_OBJECT\n\nText:\nThe discovery of the asteroid 3 Juno by Karl Ludwig Harding and 4 Vesta by Olbers, buttressed his hypothesis.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, THEORY, ASTRONOMICAL_OBJECT\n\nText:\nThe discovery of the asteroid 3 Juno by Karl Ludwig Harding and 4 Vesta by Olbers, buttressed his hypothesis.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_346", "text": "During their maturation in the thymus, they undergo a process called V (D) J recombination which conducts the development of T-cell receptor (TCR).", "requested_types": ["PROTEIN", "AWARD", "MISC"], "distractors": ["AWARD"], "gold": {"PROTEIN": ["T-cell receptor", "TCR"], "AWARD": [], "MISC": ["thymus", "V (D) J recombination"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, AWARD, MISC\n\nText:\nDuring their maturation in the thymus, they undergo a process called V (D) J recombination which conducts the development of T-cell receptor (TCR).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, AWARD, MISC\n\nText:\nDuring their maturation in the thymus, they undergo a process called V (D) J recombination which conducts the development of T-cell receptor (TCR).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_347", "text": "The chalcophile elements include: Silver, Arsenic, Bi, Cadmium, Cu, Gallium, Ge, Hg, Indium, Pb, Sulfur, Sb, Selenium, Sn, Tellurium, Thallium and Zn. Allaby, M. (2013).", "requested_types": ["PERSON", "DISCIPLINE", "THEORY", "CHEMICAL_ELEMENT", "SCIENTIST"], "distractors": ["SCIENTIST", "THEORY", "DISCIPLINE"], "gold": {"PERSON": ["Allaby, M."], "DISCIPLINE": [], "THEORY": [], "CHEMICAL_ELEMENT": ["Silver", "Arsenic", "Bi", "Cadmium", "Cu", "Gallium", "Ge", "Hg", "Indium", "Pb", "Sulfur", "Sb", "Selenium", "Sn", "Tellurium", "Thallium", "Zn"], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, DISCIPLINE, THEORY, CHEMICAL_ELEMENT, SCIENTIST\n\nText:\nThe chalcophile elements include: Silver, Arsenic, Bi, Cadmium, Cu, Gallium, Ge, Hg, Indium, Pb, Sulfur, Sb, Selenium, Sn, Tellurium, Thallium and Zn. Allaby, M. (2013).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, DISCIPLINE, THEORY, CHEMICAL_ELEMENT, SCIENTIST\n\nText:\nThe chalcophile elements include: Silver, Arsenic, Bi, Cadmium, Cu, Gallium, Ge, Hg, Indium, Pb, Sulfur, Sb, Selenium, Sn, Tellurium, Thallium and Zn. Allaby, M. (2013).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_348", "text": "LINE1 transposable elements have been identified as targets for ionizing radiation.", "requested_types": ["DISCIPLINE", "MISC", "CHEMICAL_ELEMENT", "ASTRONOMICAL_OBJECT"], "distractors": ["DISCIPLINE", "ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT"], "gold": {"DISCIPLINE": [], "MISC": ["LINE1", "ionizing radiation"], "CHEMICAL_ELEMENT": [], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, MISC, CHEMICAL_ELEMENT, ASTRONOMICAL_OBJECT\n\nText:\nLINE1 transposable elements have been identified as targets for ionizing radiation.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, MISC, CHEMICAL_ELEMENT, ASTRONOMICAL_OBJECT\n\nText:\nLINE1 transposable elements have been identified as targets for ionizing radiation.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_349", "text": "Staudinger, who initially wanted to become a botanist, studied chemistry at the Martin Luther University of Halle-Wittenberg, at the Technische Universität Darmstadt and at the Ludwig Maximilian University of Munich.", "requested_types": ["DISCIPLINE", "ENZYME", "UNIVERSITY", "SCIENTIST"], "distractors": ["ENZYME"], "gold": {"DISCIPLINE": ["chemistry"], "ENZYME": [], "UNIVERSITY": ["Martin Luther University of Halle-Wittenberg", "Technische Universität Darmstadt", "Ludwig Maximilian University of Munich"], "SCIENTIST": ["Staudinger"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ENZYME, UNIVERSITY, SCIENTIST\n\nText:\nStaudinger, who initially wanted to become a botanist, studied chemistry at the Martin Luther University of Halle-Wittenberg, at the Technische Universität Darmstadt and at the Ludwig Maximilian University of Munich.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ENZYME, UNIVERSITY, SCIENTIST\n\nText:\nStaudinger, who initially wanted to become a botanist, studied chemistry at the Martin Luther University of Halle-Wittenberg, at the Technische Universität Darmstadt and at the Ludwig Maximilian University of Munich.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_350", "text": "Chinese Physics Letters is a part of a small group of four journals from the Chinese Physical Society, the other three are: Communications in Theoretical Physics (in English, subtitled Chinese Physics A), Chinese Physics B (in English), and Chinese Physics C (in English).", "requested_types": ["ACADEMIC_JOURNAL", "MISC", "THEORY", "ORGANISATION", "UNIVERSITY"], "distractors": ["THEORY", "UNIVERSITY"], "gold": {"ACADEMIC_JOURNAL": ["Chinese Physics Letters", "Communications in Theoretical Physics", "Chinese Physics A", "Chinese Physics B", "Chinese Physics C"], "MISC": ["English"], "THEORY": [], "ORGANISATION": ["Chinese Physical Society"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, MISC, THEORY, ORGANISATION, UNIVERSITY\n\nText:\nChinese Physics Letters is a part of a small group of four journals from the Chinese Physical Society, the other three are: Communications in Theoretical Physics (in English, subtitled Chinese Physics A), Chinese Physics B (in English), and Chinese Physics C (in English).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, MISC, THEORY, ORGANISATION, UNIVERSITY\n\nText:\nChinese Physics Letters is a part of a small group of four journals from the Chinese Physical Society, the other three are: Communications in Theoretical Physics (in English, subtitled Chinese Physics A), Chinese Physics B (in English), and Chinese Physics C (in English).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_351", "text": "Mount Tuve in Ellsworth Land in Antarctica was named in honor of Merle Anthony Tuve.", "requested_types": ["LOCATION", "ENZYME", "ASTRONOMICAL_OBJECT", "DISCIPLINE", "SCIENTIST"], "distractors": ["ASTRONOMICAL_OBJECT", "ENZYME", "DISCIPLINE"], "gold": {"LOCATION": ["Mount Tuve", "Ellsworth Land", "Antarctica"], "ENZYME": [], "ASTRONOMICAL_OBJECT": [], "DISCIPLINE": [], "SCIENTIST": ["Merle Anthony Tuve"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ENZYME, ASTRONOMICAL_OBJECT, DISCIPLINE, SCIENTIST\n\nText:\nMount Tuve in Ellsworth Land in Antarctica was named in honor of Merle Anthony Tuve.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ENZYME, ASTRONOMICAL_OBJECT, DISCIPLINE, SCIENTIST\n\nText:\nMount Tuve in Ellsworth Land in Antarctica was named in honor of Merle Anthony Tuve.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_352", "text": "Supercritical fluids occur in the atmospheres of the gas giant s Jupiter and Saturn, and probably in those of the ice giant s Uranus and Neptune.", "requested_types": ["AWARD", "ASTRONOMICAL_OBJECT", "THEORY"], "distractors": ["AWARD", "THEORY"], "gold": {"AWARD": [], "ASTRONOMICAL_OBJECT": ["Jupiter", "Saturn", "Uranus", "Neptune"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ASTRONOMICAL_OBJECT, THEORY\n\nText:\nSupercritical fluids occur in the atmospheres of the gas giant s Jupiter and Saturn, and probably in those of the ice giant s Uranus and Neptune.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ASTRONOMICAL_OBJECT, THEORY\n\nText:\nSupercritical fluids occur in the atmospheres of the gas giant s Jupiter and Saturn, and probably in those of the ice giant s Uranus and Neptune.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_353", "text": "Confirmed transiting hot Jupiters that have orbital periods of less than one day include WASP-18b, WASP-19b, WASP-43b, and WASP-103b.", "requested_types": ["AWARD", "ASTRONOMICAL_OBJECT"], "distractors": ["AWARD"], "gold": {"AWARD": [], "ASTRONOMICAL_OBJECT": ["Jupiters", "WASP-18b", "WASP-19b", "WASP-43b", "WASP-103b"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ASTRONOMICAL_OBJECT\n\nText:\nConfirmed transiting hot Jupiters that have orbital periods of less than one day include WASP-18b, WASP-19b, WASP-43b, and WASP-103b.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ASTRONOMICAL_OBJECT\n\nText:\nConfirmed transiting hot Jupiters that have orbital periods of less than one day include WASP-18b, WASP-19b, WASP-43b, and WASP-103b.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_354", "text": "Fraser has been an editor for journals mBio, Journal of Bacteriology, Microbial Genomics (journal), Molecular Case Studies, and DNA and Cell Biology.", "requested_types": ["CHEMICAL_ELEMENT", "PERSON", "ACADEMIC_JOURNAL", "EVENT"], "distractors": ["EVENT", "CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_ELEMENT": [], "PERSON": ["Fraser"], "ACADEMIC_JOURNAL": ["mBio", "Journal of Bacteriology", "Microbial Genomics", "Molecular Case Studies", "DNA and Cell Biology"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, PERSON, ACADEMIC_JOURNAL, EVENT\n\nText:\nFraser has been an editor for journals mBio, Journal of Bacteriology, Microbial Genomics (journal), Molecular Case Studies, and DNA and Cell Biology.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, PERSON, ACADEMIC_JOURNAL, EVENT\n\nText:\nFraser has been an editor for journals mBio, Journal of Bacteriology, Microbial Genomics (journal), Molecular Case Studies, and DNA and Cell Biology.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_355", "text": "Aluminium hydroxide, ferrous hydroxide, Hydroxide,", "requested_types": ["LOCATION", "MISC", "CHEMICAL_COMPOUND"], "distractors": ["LOCATION", "MISC"], "gold": {"LOCATION": [], "MISC": [], "CHEMICAL_COMPOUND": ["Aluminium hydroxide", "ferrous hydroxide", "Hydroxide"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, MISC, CHEMICAL_COMPOUND\n\nText:\nAluminium hydroxide, ferrous hydroxide, Hydroxide,\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, MISC, CHEMICAL_COMPOUND\n\nText:\nAluminium hydroxide, ferrous hydroxide, Hydroxide,\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_356", "text": "The gamma pseudogene contains an inverted long interspersed nuclear element.", "requested_types": ["MISC", "ACADEMIC_JOURNAL", "ENZYME", "LOCATION"], "distractors": ["LOCATION", "ACADEMIC_JOURNAL", "ENZYME"], "gold": {"MISC": ["gamma pseudogene", "long interspersed nuclear element"], "ACADEMIC_JOURNAL": [], "ENZYME": [], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ACADEMIC_JOURNAL, ENZYME, LOCATION\n\nText:\nThe gamma pseudogene contains an inverted long interspersed nuclear element.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ACADEMIC_JOURNAL, ENZYME, LOCATION\n\nText:\nThe gamma pseudogene contains an inverted long interspersed nuclear element.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_357", "text": "Juno was originally considered a planet, along with 1 Ceres, 2 Pallas, and 4 Vesta.", "requested_types": ["EVENT", "ORGANISATION", "PROTEIN", "ASTRONOMICAL_OBJECT"], "distractors": ["ORGANISATION", "PROTEIN", "EVENT"], "gold": {"EVENT": [], "ORGANISATION": [], "PROTEIN": [], "ASTRONOMICAL_OBJECT": ["Juno", "1 Ceres", "2 Pallas", "4 Vesta"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, ORGANISATION, PROTEIN, ASTRONOMICAL_OBJECT\n\nText:\nJuno was originally considered a planet, along with 1 Ceres, 2 Pallas, and 4 Vesta.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, ORGANISATION, PROTEIN, ASTRONOMICAL_OBJECT\n\nText:\nJuno was originally considered a planet, along with 1 Ceres, 2 Pallas, and 4 Vesta.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_358", "text": "Henry Taube, Ph.D, M.Sc, B.Sc., Fellow of the Royal Society of Canada (November 30, 1915 & amp; ndash; November 16, 2005) was a Canadian -born American chemist noted for having been awarded the 1983 Nobel Prize in Chemistry for his work in the mechanisms of electron-transfer reactions, especially in metal complexes.", "requested_types": ["SCIENTIST", "LOCATION", "AWARD", "MISC", "COUNTRY"], "distractors": ["COUNTRY", "LOCATION"], "gold": {"SCIENTIST": ["Henry", "Taube"], "LOCATION": [], "AWARD": ["Fellow of the Royal Society of Canada", "Nobel Prize in Chemistry"], "MISC": ["Canadian", "American"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, LOCATION, AWARD, MISC, COUNTRY\n\nText:\nHenry Taube, Ph.D, M.Sc, B.Sc., Fellow of the Royal Society of Canada (November 30, 1915 & amp; ndash; November 16, 2005) was a Canadian -born American chemist noted for having been awarded the 1983 Nobel Prize in Chemistry for his work in the mechanisms of electron-transfer reactions, especially in metal complexes.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, LOCATION, AWARD, MISC, COUNTRY\n\nText:\nHenry Taube, Ph.D, M.Sc, B.Sc., Fellow of the Royal Society of Canada (November 30, 1915 & amp; ndash; November 16, 2005) was a Canadian -born American chemist noted for having been awarded the 1983 Nobel Prize in Chemistry for his work in the mechanisms of electron-transfer reactions, especially in metal complexes.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_359", "text": "The International Astronomical Union names all colles (small hills) on Saturn 's moon Titan after characters in Tolkien's work.", "requested_types": ["PERSON", "ORGANISATION", "ACADEMIC_JOURNAL", "COUNTRY", "MISC", "ASTRONOMICAL_OBJECT"], "distractors": ["MISC", "ACADEMIC_JOURNAL", "COUNTRY"], "gold": {"PERSON": ["Tolkien"], "ORGANISATION": ["International Astronomical Union"], "ACADEMIC_JOURNAL": [], "COUNTRY": [], "MISC": [], "ASTRONOMICAL_OBJECT": ["Saturn", "Titan"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION, ACADEMIC_JOURNAL, COUNTRY, MISC, ASTRONOMICAL_OBJECT\n\nText:\nThe International Astronomical Union names all colles (small hills) on Saturn 's moon Titan after characters in Tolkien's work.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION, ACADEMIC_JOURNAL, COUNTRY, MISC, ASTRONOMICAL_OBJECT\n\nText:\nThe International Astronomical Union names all colles (small hills) on Saturn 's moon Titan after characters in Tolkien's work.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_360", "text": "In recognition of the vast amount of scientific information amassed by both rovers, two asteroid s have been named in their honor: 37452 Spirit and 39382 Opportunity.", "requested_types": ["CHEMICAL_ELEMENT", "MISC", "ASTRONOMICAL_OBJECT"], "distractors": ["MISC", "CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_ELEMENT": [], "MISC": [], "ASTRONOMICAL_OBJECT": ["37452 Spirit", "39382 Opportunity"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, MISC, ASTRONOMICAL_OBJECT\n\nText:\nIn recognition of the vast amount of scientific information amassed by both rovers, two asteroid s have been named in their honor: 37452 Spirit and 39382 Opportunity.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, MISC, ASTRONOMICAL_OBJECT\n\nText:\nIn recognition of the vast amount of scientific information amassed by both rovers, two asteroid s have been named in their honor: 37452 Spirit and 39382 Opportunity.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_361", "text": "He has been published in a range of Academic journal s, such as the Cambridge Journal of Economics, the Oxford Bulletin of Economics and Statistics and the Oxford Review of Economic Policy.", "requested_types": ["ACADEMIC_JOURNAL", "LOCATION"], "distractors": ["LOCATION"], "gold": {"ACADEMIC_JOURNAL": ["Cambridge Journal of Economics", "Oxford Bulletin of Economics and Statistics", "Oxford Review of Economic Policy"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, LOCATION\n\nText:\nHe has been published in a range of Academic journal s, such as the Cambridge Journal of Economics, the Oxford Bulletin of Economics and Statistics and the Oxford Review of Economic Policy.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, LOCATION\n\nText:\nHe has been published in a range of Academic journal s, such as the Cambridge Journal of Economics, the Oxford Bulletin of Economics and Statistics and the Oxford Review of Economic Policy.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_362", "text": "Under typical dark sky conditions Uranus (magnitude + 5.8) can be seen as well with averted vision, as can the asteroid 4 Vesta at its brighter oppositions.", "requested_types": ["ASTRONOMICAL_OBJECT", "PROTEIN"], "distractors": ["PROTEIN"], "gold": {"ASTRONOMICAL_OBJECT": ["Uranus", "4 Vesta"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, PROTEIN\n\nText:\nUnder typical dark sky conditions Uranus (magnitude + 5.8) can be seen as well with averted vision, as can the asteroid 4 Vesta at its brighter oppositions.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, PROTEIN\n\nText:\nUnder typical dark sky conditions Uranus (magnitude + 5.8) can be seen as well with averted vision, as can the asteroid 4 Vesta at its brighter oppositions.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_363", "text": "This methylation causes other factors like Histone deacetylase (HDACs) to bind to the chromosome and propagate heterochromatin formation, even into active gene regions.", "requested_types": ["AWARD", "ENZYME", "MISC"], "distractors": ["AWARD"], "gold": {"AWARD": [], "ENZYME": ["Histone deacetylase", "HDACs"], "MISC": ["methylation", "chromosome", "heterochromatin formation"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ENZYME, MISC\n\nText:\nThis methylation causes other factors like Histone deacetylase (HDACs) to bind to the chromosome and propagate heterochromatin formation, even into active gene regions.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, ENZYME, MISC\n\nText:\nThis methylation causes other factors like Histone deacetylase (HDACs) to bind to the chromosome and propagate heterochromatin formation, even into active gene regions.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_364", "text": "The 2nd group of anions consist of Chloride, Bromide, Iodide, Nitrate.", "requested_types": ["PROTEIN", "THEORY", "MISC", "CHEMICAL_COMPOUND"], "distractors": ["PROTEIN", "MISC", "THEORY"], "gold": {"PROTEIN": [], "THEORY": [], "MISC": [], "CHEMICAL_COMPOUND": ["Chloride", "Bromide", "Iodide", "Nitrate"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, THEORY, MISC, CHEMICAL_COMPOUND\n\nText:\nThe 2nd group of anions consist of Chloride, Bromide, Iodide, Nitrate.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, THEORY, MISC, CHEMICAL_COMPOUND\n\nText:\nThe 2nd group of anions consist of Chloride, Bromide, Iodide, Nitrate.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_365", "text": "In 1987, he discovered the DNA sequence of CRISPR,", "requested_types": ["ASTRONOMICAL_OBJECT", "DISCIPLINE", "MISC", "COUNTRY"], "distractors": ["ASTRONOMICAL_OBJECT", "DISCIPLINE", "COUNTRY"], "gold": {"ASTRONOMICAL_OBJECT": [], "DISCIPLINE": [], "MISC": ["DNA", "CRISPR"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, DISCIPLINE, MISC, COUNTRY\n\nText:\nIn 1987, he discovered the DNA sequence of CRISPR,\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, DISCIPLINE, MISC, COUNTRY\n\nText:\nIn 1987, he discovered the DNA sequence of CRISPR,\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_366", "text": "It works as a transmitter from and to the tight junction, because of its association with signaling molecules (Phosphoinositide 3-kinase, Protein kinase C, YES, Protein phosphatase 2, 1).", "requested_types": ["PROTEIN", "ENZYME", "EVENT"], "distractors": ["PROTEIN", "EVENT"], "gold": {"PROTEIN": [], "ENZYME": ["Phosphoinositide 3-kinase", "Protein kinase C", "YES", "Protein phosphatase 2, 1"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ENZYME, EVENT\n\nText:\nIt works as a transmitter from and to the tight junction, because of its association with signaling molecules (Phosphoinositide 3-kinase, Protein kinase C, YES, Protein phosphatase 2, 1).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ENZYME, EVENT\n\nText:\nIt works as a transmitter from and to the tight junction, because of its association with signaling molecules (Phosphoinositide 3-kinase, Protein kinase C, YES, Protein phosphatase 2, 1).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_367", "text": "Other enzymes such as (Transglutaminase) control chromatin remodeling through proteins such as sirtuin1 (SIRT1).", "requested_types": ["PROTEIN", "ENZYME", "CHEMICAL_COMPOUND", "ASTRONOMICAL_OBJECT", "PERSON", "MISC"], "distractors": ["CHEMICAL_COMPOUND", "PERSON", "ASTRONOMICAL_OBJECT"], "gold": {"PROTEIN": ["sirtuin1", "SIRT1"], "ENZYME": ["Transglutaminase"], "CHEMICAL_COMPOUND": [], "ASTRONOMICAL_OBJECT": [], "PERSON": [], "MISC": ["chromatin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ENZYME, CHEMICAL_COMPOUND, ASTRONOMICAL_OBJECT, PERSON, MISC\n\nText:\nOther enzymes such as (Transglutaminase) control chromatin remodeling through proteins such as sirtuin1 (SIRT1).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ENZYME, CHEMICAL_COMPOUND, ASTRONOMICAL_OBJECT, PERSON, MISC\n\nText:\nOther enzymes such as (Transglutaminase) control chromatin remodeling through proteins such as sirtuin1 (SIRT1).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_368", "text": "However, various histone modifications are placed by epigenetic modifiers such as DNA methyltransferase in neurons and these marks regulate gene expression throughout the life span of the neuron.", "requested_types": ["COUNTRY", "MISC", "PROTEIN"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "MISC": ["epigenetic modifiers", "gene expression", "neuron"], "PROTEIN": ["DNA methyltransferase"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, MISC, PROTEIN\n\nText:\nHowever, various histone modifications are placed by epigenetic modifiers such as DNA methyltransferase in neurons and these marks regulate gene expression throughout the life span of the neuron.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, MISC, PROTEIN\n\nText:\nHowever, various histone modifications are placed by epigenetic modifiers such as DNA methyltransferase in neurons and these marks regulate gene expression throughout the life span of the neuron.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_369", "text": "Juno can reach + 7.5 at a favourable opposition, which is brighter than Neptune or Titan, and is the reason for it being discovered before the larger asteroids 10 Hygiea, 52 Europa, 511 Davida, and 704 Interamnia.", "requested_types": ["UNIVERSITY", "ASTRONOMICAL_OBJECT"], "distractors": ["UNIVERSITY"], "gold": {"UNIVERSITY": [], "ASTRONOMICAL_OBJECT": ["Juno", "Neptune", "Titan", "10 Hygiea", "52 Europa", "511 Davida", "704 Interamnia"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ASTRONOMICAL_OBJECT\n\nText:\nJuno can reach + 7.5 at a favourable opposition, which is brighter than Neptune or Titan, and is the reason for it being discovered before the larger asteroids 10 Hygiea, 52 Europa, 511 Davida, and 704 Interamnia.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ASTRONOMICAL_OBJECT\n\nText:\nJuno can reach + 7.5 at a favourable opposition, which is brighter than Neptune or Titan, and is the reason for it being discovered before the larger asteroids 10 Hygiea, 52 Europa, 511 Davida, and 704 Interamnia.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_370", "text": "She also played at 1986, 1989, 1993, 1995 AFC Championship, 1990, Football at the 1994 Asian Games.", "requested_types": ["CHEMICAL_ELEMENT", "MISC", "ASTRONOMICAL_OBJECT", "EVENT", "COUNTRY"], "distractors": ["COUNTRY", "ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_ELEMENT": [], "MISC": ["Football"], "ASTRONOMICAL_OBJECT": [], "EVENT": ["AFC Championship", "Asian Games"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, MISC, ASTRONOMICAL_OBJECT, EVENT, COUNTRY\n\nText:\nShe also played at 1986, 1989, 1993, 1995 AFC Championship, 1990, Football at the 1994 Asian Games.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, MISC, ASTRONOMICAL_OBJECT, EVENT, COUNTRY\n\nText:\nShe also played at 1986, 1989, 1993, 1995 AFC Championship, 1990, Football at the 1994 Asian Games.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_371", "text": "The biological effects of estrogens (and xenoestrogens) are mediated through the Estrogen receptor.", "requested_types": ["CHEMICAL_COMPOUND", "PROTEIN", "THEORY"], "distractors": ["THEORY"], "gold": {"CHEMICAL_COMPOUND": ["estrogens", "xenoestrogens"], "PROTEIN": ["Estrogen receptor"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, PROTEIN, THEORY\n\nText:\nThe biological effects of estrogens (and xenoestrogens) are mediated through the Estrogen receptor.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, PROTEIN, THEORY\n\nText:\nThe biological effects of estrogens (and xenoestrogens) are mediated through the Estrogen receptor.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_372", "text": "Pauli spent a year at the University of Göttingen as the assistant to Max Born, and the following year at the Institute for Theoretical Physics in University of Copenhagen, which later became the Niels Bohr Institute in 1965.", "requested_types": ["UNIVERSITY", "ASTRONOMICAL_OBJECT", "DISCIPLINE", "ORGANISATION", "SCIENTIST"], "distractors": ["ASTRONOMICAL_OBJECT", "DISCIPLINE"], "gold": {"UNIVERSITY": ["University of Göttingen", "University of Copenhagen"], "ASTRONOMICAL_OBJECT": [], "DISCIPLINE": [], "ORGANISATION": ["Institute for Theoretical Physics", "Niels Bohr Institute"], "SCIENTIST": ["Pauli", "Max Born"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ASTRONOMICAL_OBJECT, DISCIPLINE, ORGANISATION, SCIENTIST\n\nText:\nPauli spent a year at the University of Göttingen as the assistant to Max Born, and the following year at the Institute for Theoretical Physics in University of Copenhagen, which later became the Niels Bohr Institute in 1965.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ASTRONOMICAL_OBJECT, DISCIPLINE, ORGANISATION, SCIENTIST\n\nText:\nPauli spent a year at the University of Göttingen as the assistant to Max Born, and the following year at the Institute for Theoretical Physics in University of Copenhagen, which later became the Niels Bohr Institute in 1965.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_373", "text": "Ellen DeGeneres also became the first actress to win an award for voice acting, and the third person to do so after Robin Williams and Scott Weinger in 1992, both for Aladdin.", "requested_types": ["COUNTRY", "CHEMICAL_COMPOUND", "MISC", "ORGANISATION", "PERSON"], "distractors": ["ORGANISATION", "CHEMICAL_COMPOUND", "COUNTRY"], "gold": {"COUNTRY": [], "CHEMICAL_COMPOUND": [], "MISC": ["Aladdin"], "ORGANISATION": [], "PERSON": ["Ellen DeGeneres", "Robin Williams", "Scott Weinger"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, CHEMICAL_COMPOUND, MISC, ORGANISATION, PERSON\n\nText:\nEllen DeGeneres also became the first actress to win an award for voice acting, and the third person to do so after Robin Williams and Scott Weinger in 1992, both for Aladdin.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, CHEMICAL_COMPOUND, MISC, ORGANISATION, PERSON\n\nText:\nEllen DeGeneres also became the first actress to win an award for voice acting, and the third person to do so after Robin Williams and Scott Weinger in 1992, both for Aladdin.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_374", "text": "Assisted by activists from Southern Christian Leadership Conference, Congress of Racial Equality, SNCC, and the NAACP, African Americans and supporters took a stand to fight segregation through nonviolence.", "requested_types": ["ORGANISATION", "EVENT"], "distractors": ["EVENT"], "gold": {"ORGANISATION": ["Southern Christian Leadership Conference", "Congress of Racial Equality", "SNCC", "NAACP"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, EVENT\n\nText:\nAssisted by activists from Southern Christian Leadership Conference, Congress of Racial Equality, SNCC, and the NAACP, African Americans and supporters took a stand to fight segregation through nonviolence.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, EVENT\n\nText:\nAssisted by activists from Southern Christian Leadership Conference, Congress of Racial Equality, SNCC, and the NAACP, African Americans and supporters took a stand to fight segregation through nonviolence.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_375", "text": "DSBs can be artificially induced using genome editing technologies such as CRISPR or TALEN.", "requested_types": ["PERSON", "ORGANISATION", "MISC"], "distractors": ["ORGANISATION", "PERSON"], "gold": {"PERSON": [], "ORGANISATION": [], "MISC": ["DSBs", "CRISPR", "TALEN"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION, MISC\n\nText:\nDSBs can be artificially induced using genome editing technologies such as CRISPR or TALEN.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION, MISC\n\nText:\nDSBs can be artificially induced using genome editing technologies such as CRISPR or TALEN.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_376", "text": "Beginning in 1882, data for Mercury, Uranus, and Neptune was added to the first part, with eclipses, occultations, and other phenomena forming a separate third part.", "requested_types": ["CHEMICAL_COMPOUND", "ASTRONOMICAL_OBJECT", "UNIVERSITY"], "distractors": ["UNIVERSITY", "CHEMICAL_COMPOUND"], "gold": {"CHEMICAL_COMPOUND": [], "ASTRONOMICAL_OBJECT": ["Mercury", "Uranus", "Neptune"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ASTRONOMICAL_OBJECT, UNIVERSITY\n\nText:\nBeginning in 1882, data for Mercury, Uranus, and Neptune was added to the first part, with eclipses, occultations, and other phenomena forming a separate third part.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ASTRONOMICAL_OBJECT, UNIVERSITY\n\nText:\nBeginning in 1882, data for Mercury, Uranus, and Neptune was added to the first part, with eclipses, occultations, and other phenomena forming a separate third part.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_377", "text": "In 1966 Maiman received the American Physical Society 's Oliver E. Buckley Condensed Matter Prize and the Hertz Foundation Award for distinguished contribution in the field of science,", "requested_types": ["THEORY", "ORGANISATION", "CHEMICAL_COMPOUND", "AWARD", "SCIENTIST", "UNIVERSITY"], "distractors": ["THEORY", "CHEMICAL_COMPOUND", "UNIVERSITY"], "gold": {"THEORY": [], "ORGANISATION": ["American Physical Society"], "CHEMICAL_COMPOUND": [], "AWARD": ["Oliver E. Buckley Condensed Matter Prize", "Hertz Foundation Award"], "SCIENTIST": ["Maiman"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ORGANISATION, CHEMICAL_COMPOUND, AWARD, SCIENTIST, UNIVERSITY\n\nText:\nIn 1966 Maiman received the American Physical Society 's Oliver E. Buckley Condensed Matter Prize and the Hertz Foundation Award for distinguished contribution in the field of science,\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ORGANISATION, CHEMICAL_COMPOUND, AWARD, SCIENTIST, UNIVERSITY\n\nText:\nIn 1966 Maiman received the American Physical Society 's Oliver E. Buckley Condensed Matter Prize and the Hertz Foundation Award for distinguished contribution in the field of science,\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_378", "text": "In 1968 and 1970, Roger Penrose, Stephen Hawking, and George F. R. Ellis published papers where they showed that mathematical singularities were an inevitable initial condition of relativistic models of the Big Bang.", "requested_types": ["SCIENTIST", "THEORY", "AWARD", "CHEMICAL_ELEMENT", "ACADEMIC_JOURNAL"], "distractors": ["CHEMICAL_ELEMENT", "AWARD", "ACADEMIC_JOURNAL"], "gold": {"SCIENTIST": ["Roger Penrose", "Stephen Hawking", "George F. R. Ellis"], "THEORY": ["Big Bang"], "AWARD": [], "CHEMICAL_ELEMENT": [], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, THEORY, AWARD, CHEMICAL_ELEMENT, ACADEMIC_JOURNAL\n\nText:\nIn 1968 and 1970, Roger Penrose, Stephen Hawking, and George F. R. Ellis published papers where they showed that mathematical singularities were an inevitable initial condition of relativistic models of the Big Bang.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, THEORY, AWARD, CHEMICAL_ELEMENT, ACADEMIC_JOURNAL\n\nText:\nIn 1968 and 1970, Roger Penrose, Stephen Hawking, and George F. R. Ellis published papers where they showed that mathematical singularities were an inevitable initial condition of relativistic models of the Big Bang.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_379", "text": "He received a Doctorate, honoris causa, from the University of Paris and honorary memberships in the Royal Society, Royal Society of Edinburgh, the Indian Academy of Sciences, the International Academy of Philosophy and Sciences, and the Pontifical Academy of Sciences.", "requested_types": ["UNIVERSITY", "ORGANISATION", "COUNTRY", "MISC"], "distractors": ["COUNTRY"], "gold": {"UNIVERSITY": ["University of Paris", "Indian Academy of Sciences"], "ORGANISATION": ["Royal Society", "Royal Society of Edinburgh", "International Academy of Philosophy and Sciences", "Pontifical Academy of Sciences"], "COUNTRY": [], "MISC": ["Doctorate", "honoris causa"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ORGANISATION, COUNTRY, MISC\n\nText:\nHe received a Doctorate, honoris causa, from the University of Paris and honorary memberships in the Royal Society, Royal Society of Edinburgh, the Indian Academy of Sciences, the International Academy of Philosophy and Sciences, and the Pontifical Academy of Sciences.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ORGANISATION, COUNTRY, MISC\n\nText:\nHe received a Doctorate, honoris causa, from the University of Paris and honorary memberships in the Royal Society, Royal Society of Edinburgh, the Indian Academy of Sciences, the International Academy of Philosophy and Sciences, and the Pontifical Academy of Sciences.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_380", "text": "There are two voltage-gated calcium channels within cardiac muscle: L-type calcium channel s ('L' for Long-lasting) and T-type calcium channel s ('T' for Transient, i.e. short).", "requested_types": ["THEORY", "ENZYME", "MISC"], "distractors": ["ENZYME", "THEORY"], "gold": {"THEORY": [], "ENZYME": [], "MISC": ["voltage-gated calcium channels", "L-type calcium channel", "T-type calcium channel"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ENZYME, MISC\n\nText:\nThere are two voltage-gated calcium channels within cardiac muscle: L-type calcium channel s ('L' for Long-lasting) and T-type calcium channel s ('T' for Transient, i.e. short).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ENZYME, MISC\n\nText:\nThere are two voltage-gated calcium channels within cardiac muscle: L-type calcium channel s ('L' for Long-lasting) and T-type calcium channel s ('T' for Transient, i.e. short).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_381", "text": "In 1855, he received the Copley Medal of the Royal Society for his' very remarkable experimental researches'.", "requested_types": ["PROTEIN", "ORGANISATION", "AWARD"], "distractors": ["PROTEIN"], "gold": {"PROTEIN": [], "ORGANISATION": ["Royal Society"], "AWARD": ["Copley Medal"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ORGANISATION, AWARD\n\nText:\nIn 1855, he received the Copley Medal of the Royal Society for his' very remarkable experimental researches'.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ORGANISATION, AWARD\n\nText:\nIn 1855, he received the Copley Medal of the Royal Society for his' very remarkable experimental researches'.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_382", "text": "In chemistry, Erwin Schrödinger, Linus Pauling, Robert S. Mulliken and others noted that the consequence of Heisenberg's relation was that the electron, as a wave packet, could not be considered to have an exact location in its orbital.", "requested_types": ["SCIENTIST", "UNIVERSITY", "ACADEMIC_JOURNAL", "THEORY", "DISCIPLINE"], "distractors": ["UNIVERSITY", "ACADEMIC_JOURNAL"], "gold": {"SCIENTIST": ["Erwin Schrödinger", "Linus Pauling", "Robert S. Mulliken"], "UNIVERSITY": [], "ACADEMIC_JOURNAL": [], "THEORY": ["Heisenberg's relation"], "DISCIPLINE": ["chemistry"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, UNIVERSITY, ACADEMIC_JOURNAL, THEORY, DISCIPLINE\n\nText:\nIn chemistry, Erwin Schrödinger, Linus Pauling, Robert S. Mulliken and others noted that the consequence of Heisenberg's relation was that the electron, as a wave packet, could not be considered to have an exact location in its orbital.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, UNIVERSITY, ACADEMIC_JOURNAL, THEORY, DISCIPLINE\n\nText:\nIn chemistry, Erwin Schrödinger, Linus Pauling, Robert S. Mulliken and others noted that the consequence of Heisenberg's relation was that the electron, as a wave packet, could not be considered to have an exact location in its orbital.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_383", "text": "After 3 years at King's College London (contemporary with Rosalind Franklin) she moved to the University of Sheffield in 1955 as a demonstrator in the Biochemistry department (now Molecular Biology and Biotechnology), obtaining an MRC grant to study the iron storage protein Ferritin, publishing preliminary X-ray diffraction data in the 1st volume of the Journal of Molecular Biology in 1959.", "requested_types": ["SCIENTIST", "ORGANISATION", "THEORY", "DISCIPLINE", "PROTEIN", "ENZYME", "UNIVERSITY", "MISC", "ACADEMIC_JOURNAL", "ASTRONOMICAL_OBJECT"], "distractors": ["ENZYME", "THEORY", "ASTRONOMICAL_OBJECT"], "gold": {"SCIENTIST": ["Rosalind Franklin"], "ORGANISATION": ["MRC"], "THEORY": [], "DISCIPLINE": ["Biochemistry", "Molecular Biology", "Biotechnology"], "PROTEIN": ["Ferritin"], "ENZYME": [], "UNIVERSITY": ["King's College London", "University of Sheffield"], "MISC": ["X-ray"], "ACADEMIC_JOURNAL": ["Journal of Molecular Biology"], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ORGANISATION, THEORY, DISCIPLINE, PROTEIN, ENZYME, UNIVERSITY, MISC, ACADEMIC_JOURNAL, ASTRONOMICAL_OBJECT\n\nText:\nAfter 3 years at King's College London (contemporary with Rosalind Franklin) she moved to the University of Sheffield in 1955 as a demonstrator in the Biochemistry department (now Molecular Biology and Biotechnology), obtaining an MRC grant to study the iron storage protein Ferritin, publishing preliminary X-ray diffraction data in the 1st volume of the Journal of Molecular Biology in 1959.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ORGANISATION, THEORY, DISCIPLINE, PROTEIN, ENZYME, UNIVERSITY, MISC, ACADEMIC_JOURNAL, ASTRONOMICAL_OBJECT\n\nText:\nAfter 3 years at King's College London (contemporary with Rosalind Franklin) she moved to the University of Sheffield in 1955 as a demonstrator in the Biochemistry department (now Molecular Biology and Biotechnology), obtaining an MRC grant to study the iron storage protein Ferritin, publishing preliminary X-ray diffraction data in the 1st volume of the Journal of Molecular Biology in 1959.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_384", "text": "Several academic bodies exist to support behavior genetic research, including the Behavior Genetics Association, the International Society for Twin Studies, and the International Behavioural and Neural Genetics Society.", "requested_types": ["CHEMICAL_COMPOUND", "DISCIPLINE", "ORGANISATION", "UNIVERSITY", "EVENT"], "distractors": ["CHEMICAL_COMPOUND", "UNIVERSITY", "EVENT"], "gold": {"CHEMICAL_COMPOUND": [], "DISCIPLINE": ["behavior genetic"], "ORGANISATION": ["Behavior Genetics Association", "International Society for Twin Studies", "International Behavioural and Neural Genetics Society"], "UNIVERSITY": [], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, DISCIPLINE, ORGANISATION, UNIVERSITY, EVENT\n\nText:\nSeveral academic bodies exist to support behavior genetic research, including the Behavior Genetics Association, the International Society for Twin Studies, and the International Behavioural and Neural Genetics Society.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, DISCIPLINE, ORGANISATION, UNIVERSITY, EVENT\n\nText:\nSeveral academic bodies exist to support behavior genetic research, including the Behavior Genetics Association, the International Society for Twin Studies, and the International Behavioural and Neural Genetics Society.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_385", "text": "García studied law, first at the Pontifical Catholic University of Peru -although the official records of his tenure in this university were never found- and later earning a law degree from the National University of San Marcos in 1971.", "requested_types": ["UNIVERSITY", "COUNTRY", "PERSON"], "distractors": ["COUNTRY"], "gold": {"UNIVERSITY": ["Pontifical Catholic University of Peru", "National University of San Marcos"], "COUNTRY": [], "PERSON": ["García"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, COUNTRY, PERSON\n\nText:\nGarcía studied law, first at the Pontifical Catholic University of Peru -although the official records of his tenure in this university were never found- and later earning a law degree from the National University of San Marcos in 1971.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, COUNTRY, PERSON\n\nText:\nGarcía studied law, first at the Pontifical Catholic University of Peru -although the official records of his tenure in this university were never found- and later earning a law degree from the National University of San Marcos in 1971.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_386", "text": "The claims were initially submitted to the International Journal of Astrobiology, which rejected the paper, but were eventually published by the fringe Journal of Cosmology.", "requested_types": ["SCIENTIST", "ACADEMIC_JOURNAL", "UNIVERSITY"], "distractors": ["SCIENTIST", "UNIVERSITY"], "gold": {"SCIENTIST": [], "ACADEMIC_JOURNAL": ["International Journal of Astrobiology", "Journal of Cosmology"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ACADEMIC_JOURNAL, UNIVERSITY\n\nText:\nThe claims were initially submitted to the International Journal of Astrobiology, which rejected the paper, but were eventually published by the fringe Journal of Cosmology.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ACADEMIC_JOURNAL, UNIVERSITY\n\nText:\nThe claims were initially submitted to the International Journal of Astrobiology, which rejected the paper, but were eventually published by the fringe Journal of Cosmology.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_387", "text": "Another absolute instrumental approach was also possible with the development of light scattering theory by Einstein, Raman, Peter Debye, Bruno H. Zimm, and others.", "requested_types": ["CHEMICAL_COMPOUND", "ORGANISATION", "SCIENTIST", "DISCIPLINE", "THEORY"], "distractors": ["CHEMICAL_COMPOUND", "ORGANISATION", "DISCIPLINE"], "gold": {"CHEMICAL_COMPOUND": [], "ORGANISATION": [], "SCIENTIST": ["Einstein", "Raman", "Peter Debye", "Bruno H. Zimm"], "DISCIPLINE": [], "THEORY": ["light scattering theory"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ORGANISATION, SCIENTIST, DISCIPLINE, THEORY\n\nText:\nAnother absolute instrumental approach was also possible with the development of light scattering theory by Einstein, Raman, Peter Debye, Bruno H. Zimm, and others.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ORGANISATION, SCIENTIST, DISCIPLINE, THEORY\n\nText:\nAnother absolute instrumental approach was also possible with the development of light scattering theory by Einstein, Raman, Peter Debye, Bruno H. Zimm, and others.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_388", "text": "The invitees included Walther Bothe, Siegfried Flügge, Hans Geiger, Otto Hahn, Paul Harteck, Gerhard Hoffmann, Josef Mattauch, and Georg Stetter.", "requested_types": ["SCIENTIST", "UNIVERSITY", "CHEMICAL_COMPOUND", "PROTEIN"], "distractors": ["PROTEIN", "CHEMICAL_COMPOUND", "UNIVERSITY"], "gold": {"SCIENTIST": ["Walther Bothe", "Siegfried Flügge", "Hans Geiger", "Otto Hahn", "Paul Harteck", "Gerhard Hoffmann", "Josef Mattauch", "Georg Stetter"], "UNIVERSITY": [], "CHEMICAL_COMPOUND": [], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, UNIVERSITY, CHEMICAL_COMPOUND, PROTEIN\n\nText:\nThe invitees included Walther Bothe, Siegfried Flügge, Hans Geiger, Otto Hahn, Paul Harteck, Gerhard Hoffmann, Josef Mattauch, and Georg Stetter.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, UNIVERSITY, CHEMICAL_COMPOUND, PROTEIN\n\nText:\nThe invitees included Walther Bothe, Siegfried Flügge, Hans Geiger, Otto Hahn, Paul Harteck, Gerhard Hoffmann, Josef Mattauch, and Georg Stetter.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_389", "text": "She discovered mutations in the genes encoding the Retinoic acid receptor, the active derivative of vitamin A, in liver cancer and some types of leukemia and dissected the molecular mechanisms underlying their role in oncogenesis and treatment sensitivity.", "requested_types": ["ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND", "MISC", "PROTEIN", "PERSON"], "distractors": ["ASTRONOMICAL_OBJECT", "PERSON"], "gold": {"ASTRONOMICAL_OBJECT": [], "CHEMICAL_COMPOUND": ["vitamin A"], "MISC": ["liver cancer", "leukemia", "oncogenesis"], "PROTEIN": ["Retinoic acid receptor"], "PERSON": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, MISC, PROTEIN, PERSON\n\nText:\nShe discovered mutations in the genes encoding the Retinoic acid receptor, the active derivative of vitamin A, in liver cancer and some types of leukemia and dissected the molecular mechanisms underlying their role in oncogenesis and treatment sensitivity.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, MISC, PROTEIN, PERSON\n\nText:\nShe discovered mutations in the genes encoding the Retinoic acid receptor, the active derivative of vitamin A, in liver cancer and some types of leukemia and dissected the molecular mechanisms underlying their role in oncogenesis and treatment sensitivity.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_390", "text": "This was shown by Artur Avila and Svetlana Jitomirskaya solving the by-then famous ten martini problem with respect to the parameters).", "requested_types": ["SCIENTIST", "LOCATION", "MISC"], "distractors": ["LOCATION"], "gold": {"SCIENTIST": ["Artur Avila", "Svetlana Jitomirskaya"], "LOCATION": [], "MISC": ["ten martini problem"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, LOCATION, MISC\n\nText:\nThis was shown by Artur Avila and Svetlana Jitomirskaya solving the by-then famous ten martini problem with respect to the parameters).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, LOCATION, MISC\n\nText:\nThis was shown by Artur Avila and Svetlana Jitomirskaya solving the by-then famous ten martini problem with respect to the parameters).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_391", "text": "Uranus and 4 Vesta had most probably been seen but could not be recognized as planets because they appear so faint even at maximum brightness; Uranus' magnitude varies from + 5.3supm / sup to + 5.9supm / sup, and Vesta's from + 5.2supm / sup to + 8.5supm / sup (so that it is only visible near its opposition dates).", "requested_types": ["ACADEMIC_JOURNAL", "LOCATION", "ASTRONOMICAL_OBJECT"], "distractors": ["ACADEMIC_JOURNAL", "LOCATION"], "gold": {"ACADEMIC_JOURNAL": [], "LOCATION": [], "ASTRONOMICAL_OBJECT": ["Uranus", "4 Vesta", "Vesta"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, LOCATION, ASTRONOMICAL_OBJECT\n\nText:\nUranus and 4 Vesta had most probably been seen but could not be recognized as planets because they appear so faint even at maximum brightness; Uranus' magnitude varies from + 5.3supm / sup to + 5.9supm / sup, and Vesta's from + 5.2supm / sup to + 8.5supm / sup (so that it is only visible near its opposition dates).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, LOCATION, ASTRONOMICAL_OBJECT\n\nText:\nUranus and 4 Vesta had most probably been seen but could not be recognized as planets because they appear so faint even at maximum brightness; Uranus' magnitude varies from + 5.3supm / sup to + 5.9supm / sup, and Vesta's from + 5.2supm / sup to + 8.5supm / sup (so that it is only visible near its opposition dates).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_392", "text": "While he was there, he was among a group of physicists including Bruno Zumino, Harry Lehmann, Wolfhart Zimmermann, Kurt Symanzik, Gerhard Lüders, Reinhard Oehme, Vladimir Glaser, and Carl Friedrich von Weizsäcker.", "requested_types": ["SCIENTIST", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"SCIENTIST": ["Bruno Zumino", "Harry Lehmann", "Wolfhart Zimmermann", "Kurt Symanzik", "Gerhard Lüders", "Reinhard Oehme", "Vladimir Glaser", "Carl Friedrich von Weizsäcker"], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_ELEMENT\n\nText:\nWhile he was there, he was among a group of physicists including Bruno Zumino, Harry Lehmann, Wolfhart Zimmermann, Kurt Symanzik, Gerhard Lüders, Reinhard Oehme, Vladimir Glaser, and Carl Friedrich von Weizsäcker.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_ELEMENT\n\nText:\nWhile he was there, he was among a group of physicists including Bruno Zumino, Harry Lehmann, Wolfhart Zimmermann, Kurt Symanzik, Gerhard Lüders, Reinhard Oehme, Vladimir Glaser, and Carl Friedrich von Weizsäcker.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_393", "text": "He served on the editorial boards of several scientific journals, including NeuroImage, European Journal of Neuroscience, European Journal of Physiology, NeuroReport, Physiological Reviews, Journal of Motor Behavior, Perception, Cognition, Behavioural Brain Research, and Experimental Brain Research.", "requested_types": ["ACADEMIC_JOURNAL", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"ACADEMIC_JOURNAL": ["NeuroImage", "European Journal of Neuroscience", "European Journal of Physiology", "NeuroReport", "Physiological Reviews", "Journal of Motor Behavior", "Perception", "Cognition", "Behavioural Brain Research", "Experimental Brain Research"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, COUNTRY\n\nText:\nHe served on the editorial boards of several scientific journals, including NeuroImage, European Journal of Neuroscience, European Journal of Physiology, NeuroReport, Physiological Reviews, Journal of Motor Behavior, Perception, Cognition, Behavioural Brain Research, and Experimental Brain Research.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, COUNTRY\n\nText:\nHe served on the editorial boards of several scientific journals, including NeuroImage, European Journal of Neuroscience, European Journal of Physiology, NeuroReport, Physiological Reviews, Journal of Motor Behavior, Perception, Cognition, Behavioural Brain Research, and Experimental Brain Research.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_394", "text": "Amdahl was named an IBM Fellow in 1965, became a member of the National Academy of Engineering in 1967 and was recognized as the Centennial Alumnus of South Dakota State University in 1986.", "requested_types": ["UNIVERSITY", "ORGANISATION", "PERSON", "ACADEMIC_JOURNAL", "LOCATION", "MISC"], "distractors": ["LOCATION", "ACADEMIC_JOURNAL"], "gold": {"UNIVERSITY": ["South Dakota State University"], "ORGANISATION": ["National Academy of Engineering"], "PERSON": ["Amdahl"], "ACADEMIC_JOURNAL": [], "LOCATION": [], "MISC": ["IBM Fellow"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ORGANISATION, PERSON, ACADEMIC_JOURNAL, LOCATION, MISC\n\nText:\nAmdahl was named an IBM Fellow in 1965, became a member of the National Academy of Engineering in 1967 and was recognized as the Centennial Alumnus of South Dakota State University in 1986.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ORGANISATION, PERSON, ACADEMIC_JOURNAL, LOCATION, MISC\n\nText:\nAmdahl was named an IBM Fellow in 1965, became a member of the National Academy of Engineering in 1967 and was recognized as the Centennial Alumnus of South Dakota State University in 1986.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_395", "text": "Additionally, Histone Deacetylase A6 and A19 activity contributes to silencing of Cytochrome P450 707A and activation of NINE-CIS-EPOXYCAROTENOID DIOXYGENASE 4, 9.", "requested_types": ["MISC", "LOCATION", "ENZYME", "UNIVERSITY", "PROTEIN"], "distractors": ["MISC", "UNIVERSITY", "LOCATION"], "gold": {"MISC": [], "LOCATION": [], "ENZYME": ["Histone Deacetylase A6", "A19", "NINE-CIS-EPOXYCAROTENOID DIOXYGENASE"], "UNIVERSITY": [], "PROTEIN": ["Cytochrome P450 707A"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, LOCATION, ENZYME, UNIVERSITY, PROTEIN\n\nText:\nAdditionally, Histone Deacetylase A6 and A19 activity contributes to silencing of Cytochrome P450 707A and activation of NINE-CIS-EPOXYCAROTENOID DIOXYGENASE 4, 9.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, LOCATION, ENZYME, UNIVERSITY, PROTEIN\n\nText:\nAdditionally, Histone Deacetylase A6 and A19 activity contributes to silencing of Cytochrome P450 707A and activation of NINE-CIS-EPOXYCAROTENOID DIOXYGENASE 4, 9.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_396", "text": "Mikhail Lomonosov (1711-1765) had previously expressed similar ideas in 1748 and proved them in experiments; others whose ideas pre-date the work of Lavoisier include Jean Rey (1583-1645), Joseph Black (1728-1799), and Henry Cavendish (1731-1810).", "requested_types": ["AWARD", "SCIENTIST", "ORGANISATION"], "distractors": ["ORGANISATION", "AWARD"], "gold": {"AWARD": [], "SCIENTIST": ["Mikhail Lomonosov", "Lavoisier", "Jean Rey", "Joseph Black", "Henry Cavendish"], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, SCIENTIST, ORGANISATION\n\nText:\nMikhail Lomonosov (1711-1765) had previously expressed similar ideas in 1748 and proved them in experiments; others whose ideas pre-date the work of Lavoisier include Jean Rey (1583-1645), Joseph Black (1728-1799), and Henry Cavendish (1731-1810).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, SCIENTIST, ORGANISATION\n\nText:\nMikhail Lomonosov (1711-1765) had previously expressed similar ideas in 1748 and proved them in experiments; others whose ideas pre-date the work of Lavoisier include Jean Rey (1583-1645), Joseph Black (1728-1799), and Henry Cavendish (1731-1810).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_397", "text": "Magnetogenetic techniques involve first fusing TRPV class receptors, which are selective calcium transporters, with a paramagnetic protein (typically Ferritin).", "requested_types": ["MISC", "ACADEMIC_JOURNAL", "PROTEIN"], "distractors": ["ACADEMIC_JOURNAL"], "gold": {"MISC": ["Magnetogenetic techniques", "TRPV class receptors", "calcium transporters", "paramagnetic protein"], "ACADEMIC_JOURNAL": [], "PROTEIN": ["Ferritin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ACADEMIC_JOURNAL, PROTEIN\n\nText:\nMagnetogenetic techniques involve first fusing TRPV class receptors, which are selective calcium transporters, with a paramagnetic protein (typically Ferritin).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ACADEMIC_JOURNAL, PROTEIN\n\nText:\nMagnetogenetic techniques involve first fusing TRPV class receptors, which are selective calcium transporters, with a paramagnetic protein (typically Ferritin).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_398", "text": "The boy was sent to school at the Lycée Louis-le-Grand, part of the University of Paris, and it was decided that he would carry on the family martial tradition. The comte, the boy's great-grandfather, enrolled the boy in a program to train future Musketeers.", "requested_types": ["ORGANISATION", "UNIVERSITY", "ACADEMIC_JOURNAL", "MISC"], "distractors": ["MISC", "ACADEMIC_JOURNAL"], "gold": {"ORGANISATION": ["Lycée Louis-le-Grand"], "UNIVERSITY": ["University of Paris"], "ACADEMIC_JOURNAL": [], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, UNIVERSITY, ACADEMIC_JOURNAL, MISC\n\nText:\nThe boy was sent to school at the Lycée Louis-le-Grand, part of the University of Paris, and it was decided that he would carry on the family martial tradition. The comte, the boy's great-grandfather, enrolled the boy in a program to train future Musketeers.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, UNIVERSITY, ACADEMIC_JOURNAL, MISC\n\nText:\nThe boy was sent to school at the Lycée Louis-le-Grand, part of the University of Paris, and it was decided that he would carry on the family martial tradition. The comte, the boy's great-grandfather, enrolled the boy in a program to train future Musketeers.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_399", "text": "Mars has four known co-orbital asteroids (5261 Eureka,, all at the Lagrangian point s), and Jupiter has many (an estimated one million greater than 1 km in diameter, the Jovian trojans); there are also other small co-orbital moons in the Saturnian system: Telesto and Calypso with Tethys, and Helene and Polydeuces with Dione.", "requested_types": ["ASTRONOMICAL_OBJECT", "PROTEIN", "COUNTRY", "MISC", "EVENT"], "distractors": ["EVENT", "PROTEIN", "COUNTRY"], "gold": {"ASTRONOMICAL_OBJECT": ["Mars", "5261 Eureka", "Jupiter", "Telesto", "Calypso", "Tethys", "Helene", "Polydeuces", "Dione"], "PROTEIN": [], "COUNTRY": [], "MISC": ["Lagrangian point", "Jovian trojans", "Saturnian"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, PROTEIN, COUNTRY, MISC, EVENT\n\nText:\nMars has four known co-orbital asteroids (5261 Eureka,, all at the Lagrangian point s), and Jupiter has many (an estimated one million greater than 1 km in diameter, the Jovian trojans); there are also other small co-orbital moons in the Saturnian system: Telesto and Calypso with Tethys, and Helene and Polydeuces with Dione.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, PROTEIN, COUNTRY, MISC, EVENT\n\nText:\nMars has four known co-orbital asteroids (5261 Eureka,, all at the Lagrangian point s), and Jupiter has many (an estimated one million greater than 1 km in diameter, the Jovian trojans); there are also other small co-orbital moons in the Saturnian system: Telesto and Calypso with Tethys, and Helene and Polydeuces with Dione.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_400", "text": "A key step in nitrification is the oxidation of ammonia to hydroxylamine (NH2OH) catalyzed by the enzyme Ammonia monooxygenase (AMO) .Arciero, D.; Vannelli, T.; Logan, M.; Hopper, A. B.Degradation of trichloroethylene by the ammonia-oxidizing bacterium Nitrosomonas europaea Biochem.", "requested_types": ["PERSON", "ENZYME", "CHEMICAL_COMPOUND", "CHEMICAL_ELEMENT", "MISC"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"PERSON": [".Arciero, D.", "Vannelli, T.", "Logan, M.", "Hopper, A."], "ENZYME": ["Ammonia monooxygenase", "AMO"], "CHEMICAL_COMPOUND": ["ammonia", "hydroxylamine", "NH2OH", "trichloroethylene"], "CHEMICAL_ELEMENT": [], "MISC": ["nitrification", "oxidation", "ammonia-oxidizing bacterium Nitrosomonas europaea"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ENZYME, CHEMICAL_COMPOUND, CHEMICAL_ELEMENT, MISC\n\nText:\nA key step in nitrification is the oxidation of ammonia to hydroxylamine (NH2OH) catalyzed by the enzyme Ammonia monooxygenase (AMO) .Arciero, D.; Vannelli, T.; Logan, M.; Hopper, A. B.Degradation of trichloroethylene by the ammonia-oxidizing bacterium Nitrosomonas europaea Biochem.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ENZYME, CHEMICAL_COMPOUND, CHEMICAL_ELEMENT, MISC\n\nText:\nA key step in nitrification is the oxidation of ammonia to hydroxylamine (NH2OH) catalyzed by the enzyme Ammonia monooxygenase (AMO) .Arciero, D.; Vannelli, T.; Logan, M.; Hopper, A. B.Degradation of trichloroethylene by the ammonia-oxidizing bacterium Nitrosomonas europaea Biochem.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_401", "text": "The transform provides an alternative approach to analytic wave front set s of distributions, developed independently by the Japanese mathematicians Mikio Sato, Masaki Kashiwara and Takahiro Kawai in their approach to microlocal analysis.", "requested_types": ["SCIENTIST", "EVENT", "THEORY", "MISC"], "distractors": ["EVENT", "THEORY"], "gold": {"SCIENTIST": ["Mikio Sato", "Masaki Kashiwara", "Takahiro Kawai"], "EVENT": [], "THEORY": [], "MISC": ["wave front set", "Japanese", "microlocal analysis"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, EVENT, THEORY, MISC\n\nText:\nThe transform provides an alternative approach to analytic wave front set s of distributions, developed independently by the Japanese mathematicians Mikio Sato, Masaki Kashiwara and Takahiro Kawai in their approach to microlocal analysis.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, EVENT, THEORY, MISC\n\nText:\nThe transform provides an alternative approach to analytic wave front set s of distributions, developed independently by the Japanese mathematicians Mikio Sato, Masaki Kashiwara and Takahiro Kawai in their approach to microlocal analysis.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_402", "text": "This minor planet was named by Tadeusz Banachiewicz after Jan Śniadecki (1756-1830), a Polish professor of mathematics and astronomy, who founded the Kraków Observatory ().", "requested_types": ["CHEMICAL_COMPOUND", "PROTEIN", "MISC", "ORGANISATION", "DISCIPLINE", "SCIENTIST", "UNIVERSITY"], "distractors": ["CHEMICAL_COMPOUND", "UNIVERSITY", "PROTEIN"], "gold": {"CHEMICAL_COMPOUND": [], "PROTEIN": [], "MISC": ["minor planet", "Polish"], "ORGANISATION": ["Kraków Observatory"], "DISCIPLINE": ["mathematics", "astronomy"], "SCIENTIST": ["Tadeusz Banachiewicz", "Jan Śniadecki"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, PROTEIN, MISC, ORGANISATION, DISCIPLINE, SCIENTIST, UNIVERSITY\n\nText:\nThis minor planet was named by Tadeusz Banachiewicz after Jan Śniadecki (1756-1830), a Polish professor of mathematics and astronomy, who founded the Kraków Observatory ().\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, PROTEIN, MISC, ORGANISATION, DISCIPLINE, SCIENTIST, UNIVERSITY\n\nText:\nThis minor planet was named by Tadeusz Banachiewicz after Jan Śniadecki (1756-1830), a Polish professor of mathematics and astronomy, who founded the Kraków Observatory ().\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_403", "text": "He was elected a Fellow of the Royal Society in March 1827 and was awarded their Royal Medal the same year.", "requested_types": ["THEORY", "ASTRONOMICAL_OBJECT", "AWARD"], "distractors": ["ASTRONOMICAL_OBJECT", "THEORY"], "gold": {"THEORY": [], "ASTRONOMICAL_OBJECT": [], "AWARD": ["Fellow of the Royal Society", "Royal Medal"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ASTRONOMICAL_OBJECT, AWARD\n\nText:\nHe was elected a Fellow of the Royal Society in March 1827 and was awarded their Royal Medal the same year.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ASTRONOMICAL_OBJECT, AWARD\n\nText:\nHe was elected a Fellow of the Royal Society in March 1827 and was awarded their Royal Medal the same year.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_404", "text": "Fischer then attended the University of Bonn in 1871, but switched to the University of Strasbourg in 1872 with his study of phthaleins, and was appointed to a position at the university.", "requested_types": ["ORGANISATION", "ENZYME", "PERSON", "THEORY", "CHEMICAL_COMPOUND", "UNIVERSITY"], "distractors": ["ENZYME", "ORGANISATION", "THEORY"], "gold": {"ORGANISATION": [], "ENZYME": [], "PERSON": ["Fischer"], "THEORY": [], "CHEMICAL_COMPOUND": ["phthaleins"], "UNIVERSITY": ["University of Bonn", "University of Strasbourg"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ENZYME, PERSON, THEORY, CHEMICAL_COMPOUND, UNIVERSITY\n\nText:\nFischer then attended the University of Bonn in 1871, but switched to the University of Strasbourg in 1872 with his study of phthaleins, and was appointed to a position at the university.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ENZYME, PERSON, THEORY, CHEMICAL_COMPOUND, UNIVERSITY\n\nText:\nFischer then attended the University of Bonn in 1871, but switched to the University of Strasbourg in 1872 with his study of phthaleins, and was appointed to a position at the university.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_405", "text": "The hydroxyl radical can damage virtually all types of macromolecules: carbohydrates, nucleic acids (mutation s), lipids (lipid peroxidation), and amino acids (e.g. conversion of Phenylalanine to m - Tyrosine and o - Tyrosine).", "requested_types": ["ORGANISATION", "CHEMICAL_COMPOUND", "MISC"], "distractors": ["ORGANISATION"], "gold": {"ORGANISATION": [], "CHEMICAL_COMPOUND": ["hydroxyl radical", "carbohydrates", "nucleic acids", "lipids", "amino acids", "Phenylalanine", "m - Tyrosine", "o - Tyrosine"], "MISC": ["lipid peroxidation"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, CHEMICAL_COMPOUND, MISC\n\nText:\nThe hydroxyl radical can damage virtually all types of macromolecules: carbohydrates, nucleic acids (mutation s), lipids (lipid peroxidation), and amino acids (e.g. conversion of Phenylalanine to m - Tyrosine and o - Tyrosine).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, CHEMICAL_COMPOUND, MISC\n\nText:\nThe hydroxyl radical can damage virtually all types of macromolecules: carbohydrates, nucleic acids (mutation s), lipids (lipid peroxidation), and amino acids (e.g. conversion of Phenylalanine to m - Tyrosine and o - Tyrosine).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_406", "text": "Notable companions from the earlier series included Romana (Mary Tamm and Lalla Ward), a Time Lady; Sarah Jane Smith (Elisabeth Sladen); and Jo Grant (Katy Manning).", "requested_types": ["ORGANISATION", "PERSON", "COUNTRY", "CHEMICAL_ELEMENT"], "distractors": ["COUNTRY", "CHEMICAL_ELEMENT", "ORGANISATION"], "gold": {"ORGANISATION": [], "PERSON": ["Romana", "Mary Tamm", "Lalla Ward", "Time Lady", "Sarah Jane Smith", "Elisabeth Sladen", "Jo Grant", "Katy Manning"], "COUNTRY": [], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PERSON, COUNTRY, CHEMICAL_ELEMENT\n\nText:\nNotable companions from the earlier series included Romana (Mary Tamm and Lalla Ward), a Time Lady; Sarah Jane Smith (Elisabeth Sladen); and Jo Grant (Katy Manning).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PERSON, COUNTRY, CHEMICAL_ELEMENT\n\nText:\nNotable companions from the earlier series included Romana (Mary Tamm and Lalla Ward), a Time Lady; Sarah Jane Smith (Elisabeth Sladen); and Jo Grant (Katy Manning).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_407", "text": "Pre-implantation in the uterine wall, rate of loss is undetectable as the Human chorionic gonadotropin hormone is not secreted until implantation.", "requested_types": ["ORGANISATION", "PERSON", "MISC", "PROTEIN"], "distractors": ["PERSON", "ORGANISATION"], "gold": {"ORGANISATION": [], "PERSON": [], "MISC": ["uterine wall"], "PROTEIN": ["Human chorionic gonadotropin hormone"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PERSON, MISC, PROTEIN\n\nText:\nPre-implantation in the uterine wall, rate of loss is undetectable as the Human chorionic gonadotropin hormone is not secreted until implantation.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, PERSON, MISC, PROTEIN\n\nText:\nPre-implantation in the uterine wall, rate of loss is undetectable as the Human chorionic gonadotropin hormone is not secreted until implantation.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_408", "text": "The conclusion of the study would suggest that p53 inhibition might increase efficiency of human germline editing and that p53 function would need to be watched when developing CRISPR based therapy.", "requested_types": ["MISC", "CHEMICAL_ELEMENT", "PROTEIN"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"MISC": ["human germline editing", "CRISPR"], "CHEMICAL_ELEMENT": [], "PROTEIN": ["p53"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_ELEMENT, PROTEIN\n\nText:\nThe conclusion of the study would suggest that p53 inhibition might increase efficiency of human germline editing and that p53 function would need to be watched when developing CRISPR based therapy.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, CHEMICAL_ELEMENT, PROTEIN\n\nText:\nThe conclusion of the study would suggest that p53 inhibition might increase efficiency of human germline editing and that p53 function would need to be watched when developing CRISPR based therapy.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_409", "text": "The human and mouse genomes each contain about 30 (evolutionarily highly conserved) bHLH / PAS protein-coding genes; this gene subfamily is a subset within the 110 members of the human Basic helix-loop-helix gene superfamily.", "requested_types": ["MISC", "DISCIPLINE", "ACADEMIC_JOURNAL", "PROTEIN"], "distractors": ["ACADEMIC_JOURNAL", "MISC", "DISCIPLINE"], "gold": {"MISC": [], "DISCIPLINE": [], "ACADEMIC_JOURNAL": [], "PROTEIN": ["bHLH / PAS", "Basic helix-loop-helix"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, DISCIPLINE, ACADEMIC_JOURNAL, PROTEIN\n\nText:\nThe human and mouse genomes each contain about 30 (evolutionarily highly conserved) bHLH / PAS protein-coding genes; this gene subfamily is a subset within the 110 members of the human Basic helix-loop-helix gene superfamily.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, DISCIPLINE, ACADEMIC_JOURNAL, PROTEIN\n\nText:\nThe human and mouse genomes each contain about 30 (evolutionarily highly conserved) bHLH / PAS protein-coding genes; this gene subfamily is a subset within the 110 members of the human Basic helix-loop-helix gene superfamily.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_410", "text": "They also noticed the similarity between this test theory and Lorentz ether theory of Hendrik Lorentz, Joseph Larmor and Henri Poincaré.", "requested_types": ["THEORY", "SCIENTIST", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"THEORY": ["Lorentz ether theory"], "SCIENTIST": ["Hendrik Lorentz", "Joseph Larmor", "Henri Poincaré"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, SCIENTIST, COUNTRY\n\nText:\nThey also noticed the similarity between this test theory and Lorentz ether theory of Hendrik Lorentz, Joseph Larmor and Henri Poincaré.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, SCIENTIST, COUNTRY\n\nText:\nThey also noticed the similarity between this test theory and Lorentz ether theory of Hendrik Lorentz, Joseph Larmor and Henri Poincaré.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_411", "text": "It was founded by William Bateson and Edith Rebecca Saunders in 1919 and celebrates its centenary year in 2019.", "requested_types": ["ASTRONOMICAL_OBJECT", "MISC", "THEORY", "SCIENTIST"], "distractors": ["THEORY", "MISC", "ASTRONOMICAL_OBJECT"], "gold": {"ASTRONOMICAL_OBJECT": [], "MISC": [], "THEORY": [], "SCIENTIST": ["William Bateson", "Edith Rebecca Saunders"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, THEORY, SCIENTIST\n\nText:\nIt was founded by William Bateson and Edith Rebecca Saunders in 1919 and celebrates its centenary year in 2019.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, THEORY, SCIENTIST\n\nText:\nIt was founded by William Bateson and Edith Rebecca Saunders in 1919 and celebrates its centenary year in 2019.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_412", "text": "Supercritical fluids occur in the atmospheres of the gas giant s Jupiter and Saturn, and probably in those of the ice giant s Uranus and Neptune.", "requested_types": ["ASTRONOMICAL_OBJECT", "DISCIPLINE", "MISC"], "distractors": ["DISCIPLINE"], "gold": {"ASTRONOMICAL_OBJECT": ["Jupiter", "Saturn", "Uranus", "Neptune"], "DISCIPLINE": [], "MISC": ["Supercritical fluids"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, DISCIPLINE, MISC\n\nText:\nSupercritical fluids occur in the atmospheres of the gas giant s Jupiter and Saturn, and probably in those of the ice giant s Uranus and Neptune.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, DISCIPLINE, MISC\n\nText:\nSupercritical fluids occur in the atmospheres of the gas giant s Jupiter and Saturn, and probably in those of the ice giant s Uranus and Neptune.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_413", "text": "They include the first Albert Einstein Award (1951), the U.S. National Medal of Science (1964), honorary D.Sc. degrees from Purdue University (1961) and Harvard University (1962), and the Nature of Light Award of the U.S. National Academy of Sciences (1949).", "requested_types": ["LOCATION", "ORGANISATION", "SCIENTIST", "AWARD", "ACADEMIC_JOURNAL", "UNIVERSITY"], "distractors": ["LOCATION", "ACADEMIC_JOURNAL", "SCIENTIST"], "gold": {"LOCATION": [], "ORGANISATION": ["U.S. National Academy of Sciences"], "SCIENTIST": [], "AWARD": ["Albert Einstein Award", "U.S. National Medal of Science", "Nature of Light Award"], "ACADEMIC_JOURNAL": [], "UNIVERSITY": ["Purdue University", "Harvard University"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ORGANISATION, SCIENTIST, AWARD, ACADEMIC_JOURNAL, UNIVERSITY\n\nText:\nThey include the first Albert Einstein Award (1951), the U.S. National Medal of Science (1964), honorary D.Sc. degrees from Purdue University (1961) and Harvard University (1962), and the Nature of Light Award of the U.S. National Academy of Sciences (1949).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, ORGANISATION, SCIENTIST, AWARD, ACADEMIC_JOURNAL, UNIVERSITY\n\nText:\nThey include the first Albert Einstein Award (1951), the U.S. National Medal of Science (1964), honorary D.Sc. degrees from Purdue University (1961) and Harvard University (1962), and the Nature of Light Award of the U.S. National Academy of Sciences (1949).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_414", "text": "Szilard initially attended Budapest University of Technology and Economics in Budapest, but his engineering studies were interrupted by service in the Austro-Hungarian Army during World War I. He left Hungary for Germany in 1919, enrolling at Technical University of Berlin, but became bored with engineering and transferred to Humboldt University of Berlin, where he studied physics.", "requested_types": ["COUNTRY", "UNIVERSITY", "ORGANISATION", "THEORY", "LOCATION", "PERSON", "EVENT", "CHEMICAL_COMPOUND", "SCIENTIST", "DISCIPLINE"], "distractors": ["CHEMICAL_COMPOUND", "THEORY", "PERSON"], "gold": {"COUNTRY": ["Hungary", "Germany"], "UNIVERSITY": ["Budapest University of Technology and Economics", "Technical University of Berlin", "Humboldt University of Berlin"], "ORGANISATION": ["Austro-Hungarian Army"], "THEORY": [], "LOCATION": ["Budapest"], "PERSON": [], "EVENT": ["World War I."], "CHEMICAL_COMPOUND": [], "SCIENTIST": ["Szilard"], "DISCIPLINE": ["physics"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, UNIVERSITY, ORGANISATION, THEORY, LOCATION, PERSON, EVENT, CHEMICAL_COMPOUND, SCIENTIST, DISCIPLINE\n\nText:\nSzilard initially attended Budapest University of Technology and Economics in Budapest, but his engineering studies were interrupted by service in the Austro-Hungarian Army during World War I. He left Hungary for Germany in 1919, enrolling at Technical University of Berlin, but became bored with engineering and transferred to Humboldt University of Berlin, where he studied physics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, UNIVERSITY, ORGANISATION, THEORY, LOCATION, PERSON, EVENT, CHEMICAL_COMPOUND, SCIENTIST, DISCIPLINE\n\nText:\nSzilard initially attended Budapest University of Technology and Economics in Budapest, but his engineering studies were interrupted by service in the Austro-Hungarian Army during World War I. He left Hungary for Germany in 1919, enrolling at Technical University of Berlin, but became bored with engineering and transferred to Humboldt University of Berlin, where he studied physics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_415", "text": "Notable proponents of this approach include Arjun Appadurai, James Clifford, George E. Marcus, Sidney Mintz, Michael Taussig, Eric Wolf and Ronald Daus.", "requested_types": ["SCIENTIST", "ASTRONOMICAL_OBJECT", "PERSON"], "distractors": ["PERSON", "ASTRONOMICAL_OBJECT"], "gold": {"SCIENTIST": ["Arjun Appadurai", "James Clifford", "George E. Marcus", "Sidney Mintz", "Michael Taussig", "Eric Wolf", "Ronald Daus"], "ASTRONOMICAL_OBJECT": [], "PERSON": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ASTRONOMICAL_OBJECT, PERSON\n\nText:\nNotable proponents of this approach include Arjun Appadurai, James Clifford, George E. Marcus, Sidney Mintz, Michael Taussig, Eric Wolf and Ronald Daus.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ASTRONOMICAL_OBJECT, PERSON\n\nText:\nNotable proponents of this approach include Arjun Appadurai, James Clifford, George E. Marcus, Sidney Mintz, Michael Taussig, Eric Wolf and Ronald Daus.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_416", "text": "In 2002, it received nominations for Saturn Award for Best Actress on Television (Black), Saturn Award for Best Supporting Actor on Television (Anthony Simcoe as the Luxan warrior Ka D'Argo), and Saturn Award for Best Supporting Actress on Television (Gigi Edgley as the Nebari rogue Chiana).", "requested_types": ["UNIVERSITY", "PERSON", "AWARD"], "distractors": ["UNIVERSITY"], "gold": {"UNIVERSITY": [], "PERSON": ["Anthony Simcoe", "Ka D'Argo", "Gigi Edgley", "Chiana"], "AWARD": ["Saturn Award for Best Actress on Television", "Saturn Award for Best Supporting Actor on Television", "Saturn Award for Best Supporting Actress on Television"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, PERSON, AWARD\n\nText:\nIn 2002, it received nominations for Saturn Award for Best Actress on Television (Black), Saturn Award for Best Supporting Actor on Television (Anthony Simcoe as the Luxan warrior Ka D'Argo), and Saturn Award for Best Supporting Actress on Television (Gigi Edgley as the Nebari rogue Chiana).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, PERSON, AWARD\n\nText:\nIn 2002, it received nominations for Saturn Award for Best Actress on Television (Black), Saturn Award for Best Supporting Actor on Television (Anthony Simcoe as the Luxan warrior Ka D'Argo), and Saturn Award for Best Supporting Actress on Television (Gigi Edgley as the Nebari rogue Chiana).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_417", "text": "to University of Göttingen in 1913, to ETH Zurich in 1920, to University of Leipzig in 1927, and in 1934 to Humboldt University of Berlin, where, succeeding Einstein, he became director of the Kaiser Wilhelm Institute for Physics (now named the Max-Planck-Institut) whose facilities were built only during Debye's era.", "requested_types": ["PERSON", "UNIVERSITY", "ORGANISATION", "DISCIPLINE", "SCIENTIST"], "distractors": ["DISCIPLINE", "PERSON"], "gold": {"PERSON": [], "UNIVERSITY": ["University of Göttingen", "ETH Zurich", "University of Leipzig", "Humboldt University of Berlin"], "ORGANISATION": ["Kaiser Wilhelm Institute for Physics", "Max-Planck-Institut"], "DISCIPLINE": [], "SCIENTIST": ["Einstein", "Debye"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, UNIVERSITY, ORGANISATION, DISCIPLINE, SCIENTIST\n\nText:\nto University of Göttingen in 1913, to ETH Zurich in 1920, to University of Leipzig in 1927, and in 1934 to Humboldt University of Berlin, where, succeeding Einstein, he became director of the Kaiser Wilhelm Institute for Physics (now named the Max-Planck-Institut) whose facilities were built only during Debye's era.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, UNIVERSITY, ORGANISATION, DISCIPLINE, SCIENTIST\n\nText:\nto University of Göttingen in 1913, to ETH Zurich in 1920, to University of Leipzig in 1927, and in 1934 to Humboldt University of Berlin, where, succeeding Einstein, he became director of the Kaiser Wilhelm Institute for Physics (now named the Max-Planck-Institut) whose facilities were built only during Debye's era.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_418", "text": "She is an active voice in numerous committees of the National Academy of Sciences, the American Association for the Advancement of Science (AAAS), and the National Science Foundation.", "requested_types": ["ENZYME", "ORGANISATION"], "distractors": ["ENZYME"], "gold": {"ENZYME": [], "ORGANISATION": ["National Academy of Sciences", "American Association for the Advancement of Science", "AAAS", "National Science Foundation"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ORGANISATION\n\nText:\nShe is an active voice in numerous committees of the National Academy of Sciences, the American Association for the Advancement of Science (AAAS), and the National Science Foundation.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ORGANISATION\n\nText:\nShe is an active voice in numerous committees of the National Academy of Sciences, the American Association for the Advancement of Science (AAAS), and the National Science Foundation.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_419", "text": "East Malaysia currently has two public universities, namely Universiti Malaysia Sarawak (UNIMAS) and Universiti Malaysia Sabah (UMS).", "requested_types": ["UNIVERSITY", "DISCIPLINE", "AWARD", "LOCATION", "PERSON"], "distractors": ["PERSON", "DISCIPLINE", "AWARD"], "gold": {"UNIVERSITY": ["Universiti Malaysia Sarawak", "UNIMAS", "Universiti Malaysia Sabah", "UMS"], "DISCIPLINE": [], "AWARD": [], "LOCATION": ["East Malaysia"], "PERSON": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, DISCIPLINE, AWARD, LOCATION, PERSON\n\nText:\nEast Malaysia currently has two public universities, namely Universiti Malaysia Sarawak (UNIMAS) and Universiti Malaysia Sabah (UMS).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, DISCIPLINE, AWARD, LOCATION, PERSON\n\nText:\nEast Malaysia currently has two public universities, namely Universiti Malaysia Sarawak (UNIMAS) and Universiti Malaysia Sabah (UMS).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_420", "text": "He became an expert on the electrical conductivity of gases, the properties of ion s, and the behavior of atmospheric electricity, publishing in journals including the Physical Review, Journal of Applied Physics, Journal of Chemical Physics, and the Journal of Atmospheric Electricity and Terrestrial Magnetism.", "requested_types": ["MISC", "ACADEMIC_JOURNAL", "CHEMICAL_COMPOUND", "EVENT"], "distractors": ["MISC", "EVENT", "CHEMICAL_COMPOUND"], "gold": {"MISC": [], "ACADEMIC_JOURNAL": ["Physical Review", "Journal of Applied Physics", "Journal of Chemical Physics", "Journal of Atmospheric Electricity and Terrestrial Magnetism"], "CHEMICAL_COMPOUND": [], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ACADEMIC_JOURNAL, CHEMICAL_COMPOUND, EVENT\n\nText:\nHe became an expert on the electrical conductivity of gases, the properties of ion s, and the behavior of atmospheric electricity, publishing in journals including the Physical Review, Journal of Applied Physics, Journal of Chemical Physics, and the Journal of Atmospheric Electricity and Terrestrial Magnetism.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ACADEMIC_JOURNAL, CHEMICAL_COMPOUND, EVENT\n\nText:\nHe became an expert on the electrical conductivity of gases, the properties of ion s, and the behavior of atmospheric electricity, publishing in journals including the Physical Review, Journal of Applied Physics, Journal of Chemical Physics, and the Journal of Atmospheric Electricity and Terrestrial Magnetism.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_421", "text": "Interaction of PAMPs with PRPs (pattern-recognition proteins) activates a series of Serine protease and those proteolytically cleave the prophenoloxidase (proPO) zymogen and activate phenoxidase (PO).", "requested_types": ["ENZYME", "DISCIPLINE", "MISC"], "distractors": ["DISCIPLINE"], "gold": {"ENZYME": ["Serine protease"], "DISCIPLINE": [], "MISC": ["PAMPs", "PRPs", "pattern-recognition proteins", "prophenoloxidase", "proPO", "zymogen", "phenoxidase", "PO"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, DISCIPLINE, MISC\n\nText:\nInteraction of PAMPs with PRPs (pattern-recognition proteins) activates a series of Serine protease and those proteolytically cleave the prophenoloxidase (proPO) zymogen and activate phenoxidase (PO).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, DISCIPLINE, MISC\n\nText:\nInteraction of PAMPs with PRPs (pattern-recognition proteins) activates a series of Serine protease and those proteolytically cleave the prophenoloxidase (proPO) zymogen and activate phenoxidase (PO).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_422", "text": "His lab identified a set of transcription factor s that enabled activation of miR-430 and a large fraction of the genome after fertilization: maternal Nanog, Oct4 and SOX gene family .Nanog, Pou5f1 and SoxB1 activate zygotic gene expression during the maternal-to-zygotic transition.", "requested_types": ["MISC", "PROTEIN", "AWARD", "THEORY"], "distractors": ["AWARD", "THEORY"], "gold": {"MISC": ["transcription factor", "miR-430", "fertilization", "Nanog", "SOX gene family .Nanog", "Pou5f1", "SoxB1", "zygotic gene", "maternal-to-zygotic transition"], "PROTEIN": ["Oct4"], "AWARD": [], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, AWARD, THEORY\n\nText:\nHis lab identified a set of transcription factor s that enabled activation of miR-430 and a large fraction of the genome after fertilization: maternal Nanog, Oct4 and SOX gene family .Nanog, Pou5f1 and SoxB1 activate zygotic gene expression during the maternal-to-zygotic transition.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, AWARD, THEORY\n\nText:\nHis lab identified a set of transcription factor s that enabled activation of miR-430 and a large fraction of the genome after fertilization: maternal Nanog, Oct4 and SOX gene family .Nanog, Pou5f1 and SoxB1 activate zygotic gene expression during the maternal-to-zygotic transition.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_423", "text": "The purification of penicillin was achieved by Ernst Chain, Norman Heatley and Edward Abraham, with Chain and Abraham eventually developing a theoretical model for its chemical structure, which was later confirmed by Dorothy Hodgkin using X-ray crystallography.", "requested_types": ["PERSON", "MISC", "SCIENTIST"], "distractors": ["PERSON"], "gold": {"PERSON": [], "MISC": ["purification of penicillin", "X-ray crystallography"], "SCIENTIST": ["Ernst Chain", "Norman Heatley", "Edward Abraham", "Chain", "Abraham", "Dorothy Hodgkin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, MISC, SCIENTIST\n\nText:\nThe purification of penicillin was achieved by Ernst Chain, Norman Heatley and Edward Abraham, with Chain and Abraham eventually developing a theoretical model for its chemical structure, which was later confirmed by Dorothy Hodgkin using X-ray crystallography.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, MISC, SCIENTIST\n\nText:\nThe purification of penicillin was achieved by Ernst Chain, Norman Heatley and Edward Abraham, with Chain and Abraham eventually developing a theoretical model for its chemical structure, which was later confirmed by Dorothy Hodgkin using X-ray crystallography.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_424", "text": "Taurasi also received many personal accolades at UConn including the 2003 and 2004 Naismith College Player of the Year awards, the 2003 Wade Trophy, the 2003 and 2004 Honda Sports Award and the 2003 Associated Press Player of the Year award.", "requested_types": ["UNIVERSITY", "AWARD", "CHEMICAL_COMPOUND", "COUNTRY", "PERSON"], "distractors": ["CHEMICAL_COMPOUND", "COUNTRY"], "gold": {"UNIVERSITY": ["UConn"], "AWARD": ["Naismith College Player of the Year awards", "Wade Trophy", "Honda Sports Award", "Associated Press Player of the Year award"], "CHEMICAL_COMPOUND": [], "COUNTRY": [], "PERSON": ["Taurasi"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, AWARD, CHEMICAL_COMPOUND, COUNTRY, PERSON\n\nText:\nTaurasi also received many personal accolades at UConn including the 2003 and 2004 Naismith College Player of the Year awards, the 2003 Wade Trophy, the 2003 and 2004 Honda Sports Award and the 2003 Associated Press Player of the Year award.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, AWARD, CHEMICAL_COMPOUND, COUNTRY, PERSON\n\nText:\nTaurasi also received many personal accolades at UConn including the 2003 and 2004 Naismith College Player of the Year awards, the 2003 Wade Trophy, the 2003 and 2004 Honda Sports Award and the 2003 Associated Press Player of the Year award.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_425", "text": "On 15 June 1826 he was elected Fellow of the Royal Society, and in 1830 was one of the founders of the Royal Geographical Society (RGS).", "requested_types": ["ORGANISATION", "AWARD", "ENZYME"], "distractors": ["ENZYME"], "gold": {"ORGANISATION": ["Royal Geographical Society", "RGS"], "AWARD": ["Fellow of the Royal Society"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, AWARD, ENZYME\n\nText:\nOn 15 June 1826 he was elected Fellow of the Royal Society, and in 1830 was one of the founders of the Royal Geographical Society (RGS).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, AWARD, ENZYME\n\nText:\nOn 15 June 1826 he was elected Fellow of the Royal Society, and in 1830 was one of the founders of the Royal Geographical Society (RGS).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_426", "text": "The entire set is processed through cycles of: (a) hybridization at 60 ° C; (b) elongation via Taq polymerase and a standard ligase; and (c) denaturation at 95 ° C, forming progressively longer contiguous strands and ultimately resulting in the final genome.", "requested_types": ["MISC", "THEORY", "ORGANISATION", "ENZYME", "AWARD"], "distractors": ["THEORY", "AWARD", "ORGANISATION"], "gold": {"MISC": ["hybridization", "elongation", "denaturation", "genome"], "THEORY": [], "ORGANISATION": [], "ENZYME": ["Taq polymerase", "ligase"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, THEORY, ORGANISATION, ENZYME, AWARD\n\nText:\nThe entire set is processed through cycles of: (a) hybridization at 60 ° C; (b) elongation via Taq polymerase and a standard ligase; and (c) denaturation at 95 ° C, forming progressively longer contiguous strands and ultimately resulting in the final genome.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, THEORY, ORGANISATION, ENZYME, AWARD\n\nText:\nThe entire set is processed through cycles of: (a) hybridization at 60 ° C; (b) elongation via Taq polymerase and a standard ligase; and (c) denaturation at 95 ° C, forming progressively longer contiguous strands and ultimately resulting in the final genome.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_427", "text": "This includes Pluto, which is constrained in its orbit by the gravity of Neptune and shares its orbital neighbourhood with many Kuiper belt objects.", "requested_types": ["ORGANISATION", "ASTRONOMICAL_OBJECT", "MISC", "ENZYME", "COUNTRY"], "distractors": ["ENZYME", "ORGANISATION", "COUNTRY"], "gold": {"ORGANISATION": [], "ASTRONOMICAL_OBJECT": ["Pluto", "Neptune"], "MISC": ["Kuiper belt"], "ENZYME": [], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ASTRONOMICAL_OBJECT, MISC, ENZYME, COUNTRY\n\nText:\nThis includes Pluto, which is constrained in its orbit by the gravity of Neptune and shares its orbital neighbourhood with many Kuiper belt objects.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, ASTRONOMICAL_OBJECT, MISC, ENZYME, COUNTRY\n\nText:\nThis includes Pluto, which is constrained in its orbit by the gravity of Neptune and shares its orbital neighbourhood with many Kuiper belt objects.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_428", "text": "The importance of the predicted return based on the calculation by Encke was rewarded by the Royal Astronomical Society in London by presenting their Gold Medal of the Royal Astronomical Society to him in 1824.", "requested_types": ["MISC", "ORGANISATION", "SCIENTIST", "LOCATION", "AWARD"], "distractors": ["MISC"], "gold": {"MISC": [], "ORGANISATION": ["Royal Astronomical Society"], "SCIENTIST": ["Encke"], "LOCATION": ["London"], "AWARD": ["Gold Medal of the Royal Astronomical Society"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ORGANISATION, SCIENTIST, LOCATION, AWARD\n\nText:\nThe importance of the predicted return based on the calculation by Encke was rewarded by the Royal Astronomical Society in London by presenting their Gold Medal of the Royal Astronomical Society to him in 1824.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ORGANISATION, SCIENTIST, LOCATION, AWARD\n\nText:\nThe importance of the predicted return based on the calculation by Encke was rewarded by the Royal Astronomical Society in London by presenting their Gold Medal of the Royal Astronomical Society to him in 1824.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_429", "text": "That work has led to identification and molecular cloning of the IκB kinase (IKK) complex, which has turned out to be one of the major activators of the inflammatory response and innate immunity.", "requested_types": ["THEORY", "ENZYME", "ASTRONOMICAL_OBJECT", "MISC", "PROTEIN"], "distractors": ["ASTRONOMICAL_OBJECT", "PROTEIN", "THEORY"], "gold": {"THEORY": [], "ENZYME": ["IκB kinase", "IKK"], "ASTRONOMICAL_OBJECT": [], "MISC": ["inflammatory response", "innate immunity"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ENZYME, ASTRONOMICAL_OBJECT, MISC, PROTEIN\n\nText:\nThat work has led to identification and molecular cloning of the IκB kinase (IKK) complex, which has turned out to be one of the major activators of the inflammatory response and innate immunity.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, ENZYME, ASTRONOMICAL_OBJECT, MISC, PROTEIN\n\nText:\nThat work has led to identification and molecular cloning of the IκB kinase (IKK) complex, which has turned out to be one of the major activators of the inflammatory response and innate immunity.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_430", "text": "Purity standards for reagents are set by organizations such as ASTM International or the American Chemical Society.", "requested_types": ["ORGANISATION", "DISCIPLINE", "AWARD"], "distractors": ["DISCIPLINE", "AWARD"], "gold": {"ORGANISATION": ["ASTM International", "American Chemical Society"], "DISCIPLINE": [], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, DISCIPLINE, AWARD\n\nText:\nPurity standards for reagents are set by organizations such as ASTM International or the American Chemical Society.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, DISCIPLINE, AWARD\n\nText:\nPurity standards for reagents are set by organizations such as ASTM International or the American Chemical Society.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_431", "text": "He was later awarded the Gold Medal of the Royal Astronomical Society and the Royal Medal of the Royal Society for this work.", "requested_types": ["THEORY", "EVENT", "AWARD"], "distractors": ["EVENT", "THEORY"], "gold": {"THEORY": [], "EVENT": [], "AWARD": ["Gold Medal of the Royal Astronomical Society", "Royal Medal of the Royal Society"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, EVENT, AWARD\n\nText:\nHe was later awarded the Gold Medal of the Royal Astronomical Society and the Royal Medal of the Royal Society for this work.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, EVENT, AWARD\n\nText:\nHe was later awarded the Gold Medal of the Royal Astronomical Society and the Royal Medal of the Royal Society for this work.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_432", "text": "Early pioneers also include Alexey Grigorevich Ivakhnenko, Teuvo Kohonen, Stephen Grossberg, Kunihiko Fukushima, Christoph von der Malsburg, David Willshaw, Shun-Ichi Amari, Bernard Widrow, John Hopfield, Eduardo R. Caianiello, and others.", "requested_types": ["MISC", "ACADEMIC_JOURNAL", "LOCATION", "SCIENTIST"], "distractors": ["LOCATION", "MISC", "ACADEMIC_JOURNAL"], "gold": {"MISC": [], "ACADEMIC_JOURNAL": [], "LOCATION": [], "SCIENTIST": ["Alexey Grigorevich Ivakhnenko", "Teuvo Kohonen", "Stephen Grossberg", "Kunihiko Fukushima", "Christoph von der Malsburg", "David Willshaw", "Shun-Ichi Amari", "Bernard Widrow", "John Hopfield", "Eduardo R. Caianiello"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ACADEMIC_JOURNAL, LOCATION, SCIENTIST\n\nText:\nEarly pioneers also include Alexey Grigorevich Ivakhnenko, Teuvo Kohonen, Stephen Grossberg, Kunihiko Fukushima, Christoph von der Malsburg, David Willshaw, Shun-Ichi Amari, Bernard Widrow, John Hopfield, Eduardo R. Caianiello, and others.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ACADEMIC_JOURNAL, LOCATION, SCIENTIST\n\nText:\nEarly pioneers also include Alexey Grigorevich Ivakhnenko, Teuvo Kohonen, Stephen Grossberg, Kunihiko Fukushima, Christoph von der Malsburg, David Willshaw, Shun-Ichi Amari, Bernard Widrow, John Hopfield, Eduardo R. Caianiello, and others.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_433", "text": "The others: William Henry Bragg (1915) and William Lawrence Bragg (1915); J. J. Thomson (1906) and George Paget Thomson (1937); and Manne Siegbahn (1924) and Kai M. Siegbahn (1981).", "requested_types": ["EVENT", "CHEMICAL_COMPOUND", "SCIENTIST", "LOCATION"], "distractors": ["EVENT", "LOCATION", "CHEMICAL_COMPOUND"], "gold": {"EVENT": [], "CHEMICAL_COMPOUND": [], "SCIENTIST": ["William Henry Bragg", "William Lawrence Bragg", "J. J. Thomson", "George Paget Thomson", "Manne Siegbahn", "Kai M. Siegbahn"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, CHEMICAL_COMPOUND, SCIENTIST, LOCATION\n\nText:\nThe others: William Henry Bragg (1915) and William Lawrence Bragg (1915); J. J. Thomson (1906) and George Paget Thomson (1937); and Manne Siegbahn (1924) and Kai M. Siegbahn (1981).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, CHEMICAL_COMPOUND, SCIENTIST, LOCATION\n\nText:\nThe others: William Henry Bragg (1915) and William Lawrence Bragg (1915); J. J. Thomson (1906) and George Paget Thomson (1937); and Manne Siegbahn (1924) and Kai M. Siegbahn (1981).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_434", "text": "On Broadway, the song was originally performed by Chita Rivera, with Candy Brown, Cheryl Clark, Graciela Daniele, Michon Peacock and Pamela Sousa.", "requested_types": ["CHEMICAL_ELEMENT", "LOCATION", "ORGANISATION", "ACADEMIC_JOURNAL", "PERSON"], "distractors": ["LOCATION", "ACADEMIC_JOURNAL", "CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_ELEMENT": [], "LOCATION": [], "ORGANISATION": ["Broadway"], "ACADEMIC_JOURNAL": [], "PERSON": ["Chita Rivera", "Candy Brown", "Cheryl Clark", "Graciela Daniele", "Michon Peacock", "Pamela Sousa"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, LOCATION, ORGANISATION, ACADEMIC_JOURNAL, PERSON\n\nText:\nOn Broadway, the song was originally performed by Chita Rivera, with Candy Brown, Cheryl Clark, Graciela Daniele, Michon Peacock and Pamela Sousa.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, LOCATION, ORGANISATION, ACADEMIC_JOURNAL, PERSON\n\nText:\nOn Broadway, the song was originally performed by Chita Rivera, with Candy Brown, Cheryl Clark, Graciela Daniele, Michon Peacock and Pamela Sousa.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_435", "text": "Sydney Goldstein, who also wrote the Royal Society memoir for Kármán, reviewed the autobiography Sydney Goldstein (1968) Journal of Fluid Mechanics 33 (2) and remembered an eminent engineer and scientist, warm-hearted and witty, much traveled, well-known by many, devoted to international collaboration, who, in his own words, as a scientist found the military 'the most comfortable group to deal with'.", "requested_types": ["SCIENTIST", "ACADEMIC_JOURNAL", "ORGANISATION", "UNIVERSITY"], "distractors": ["UNIVERSITY"], "gold": {"SCIENTIST": ["Sydney Goldstein", "Kármán"], "ACADEMIC_JOURNAL": ["Journal of Fluid Mechanics"], "ORGANISATION": ["Royal Society"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ACADEMIC_JOURNAL, ORGANISATION, UNIVERSITY\n\nText:\nSydney Goldstein, who also wrote the Royal Society memoir for Kármán, reviewed the autobiography Sydney Goldstein (1968) Journal of Fluid Mechanics 33 (2) and remembered an eminent engineer and scientist, warm-hearted and witty, much traveled, well-known by many, devoted to international collaboration, who, in his own words, as a scientist found the military 'the most comfortable group to deal with'.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ACADEMIC_JOURNAL, ORGANISATION, UNIVERSITY\n\nText:\nSydney Goldstein, who also wrote the Royal Society memoir for Kármán, reviewed the autobiography Sydney Goldstein (1968) Journal of Fluid Mechanics 33 (2) and remembered an eminent engineer and scientist, warm-hearted and witty, much traveled, well-known by many, devoted to international collaboration, who, in his own words, as a scientist found the military 'the most comfortable group to deal with'.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_436", "text": "Epigenetic modifications to the genome, including Histone, DNA methylation, and the modulation of RNAi, are major epigenetic events used to modulate gene expression.", "requested_types": ["PROTEIN", "ORGANISATION", "MISC"], "distractors": ["ORGANISATION"], "gold": {"PROTEIN": ["Histone"], "ORGANISATION": [], "MISC": ["Epigenetic modifications", "DNA methylation", "RNAi"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ORGANISATION, MISC\n\nText:\nEpigenetic modifications to the genome, including Histone, DNA methylation, and the modulation of RNAi, are major epigenetic events used to modulate gene expression.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ORGANISATION, MISC\n\nText:\nEpigenetic modifications to the genome, including Histone, DNA methylation, and the modulation of RNAi, are major epigenetic events used to modulate gene expression.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_437", "text": "Confirmed transiting hot Jupiters that have orbital periods of less than one day include WASP-18b, WASP-19b, WASP-43b, and WASP-103b.", "requested_types": ["ASTRONOMICAL_OBJECT", "ACADEMIC_JOURNAL", "MISC"], "distractors": ["ACADEMIC_JOURNAL"], "gold": {"ASTRONOMICAL_OBJECT": ["WASP-18b", "WASP-19b", "WASP-43b", "WASP-103b"], "ACADEMIC_JOURNAL": [], "MISC": ["hot Jupiters"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ACADEMIC_JOURNAL, MISC\n\nText:\nConfirmed transiting hot Jupiters that have orbital periods of less than one day include WASP-18b, WASP-19b, WASP-43b, and WASP-103b.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ACADEMIC_JOURNAL, MISC\n\nText:\nConfirmed transiting hot Jupiters that have orbital periods of less than one day include WASP-18b, WASP-19b, WASP-43b, and WASP-103b.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_438", "text": "Among the honors he's received are the Federation Aeronautique Internationale 's De la Vaulx Medal in 1970 for his Apollo 9 flight, both of NASA's Distinguished Service and Exceptional Service medals, and, unusual for an astronaut, an Emmy Award from the U.S. National Academy of Television Arts and Sciences for transmitting the first live TV pictures from space.", "requested_types": ["DISCIPLINE", "AWARD", "ORGANISATION", "COUNTRY", "MISC", "ACADEMIC_JOURNAL"], "distractors": ["DISCIPLINE", "ACADEMIC_JOURNAL"], "gold": {"DISCIPLINE": [], "AWARD": ["De la Vaulx Medal", "Distinguished Service", "Exceptional Service medals", "Emmy Award"], "ORGANISATION": ["Federation Aeronautique Internationale", "NASA", "National Academy of Television Arts and Sciences"], "COUNTRY": ["U.S."], "MISC": ["Apollo 9 flight"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, AWARD, ORGANISATION, COUNTRY, MISC, ACADEMIC_JOURNAL\n\nText:\nAmong the honors he's received are the Federation Aeronautique Internationale 's De la Vaulx Medal in 1970 for his Apollo 9 flight, both of NASA's Distinguished Service and Exceptional Service medals, and, unusual for an astronaut, an Emmy Award from the U.S. National Academy of Television Arts and Sciences for transmitting the first live TV pictures from space.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, AWARD, ORGANISATION, COUNTRY, MISC, ACADEMIC_JOURNAL\n\nText:\nAmong the honors he's received are the Federation Aeronautique Internationale 's De la Vaulx Medal in 1970 for his Apollo 9 flight, both of NASA's Distinguished Service and Exceptional Service medals, and, unusual for an astronaut, an Emmy Award from the U.S. National Academy of Television Arts and Sciences for transmitting the first live TV pictures from space.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_439", "text": "Radioactivity was discovered in 1896 by Henri Becquerel in uranium, and subsequently observed by Marie Curie and Pierre Curie in thorium and in the new elements polonium and radium.", "requested_types": ["PERSON", "ORGANISATION", "CHEMICAL_COMPOUND", "SCIENTIST", "CHEMICAL_ELEMENT"], "distractors": ["ORGANISATION", "PERSON", "CHEMICAL_COMPOUND"], "gold": {"PERSON": [], "ORGANISATION": [], "CHEMICAL_COMPOUND": [], "SCIENTIST": ["Henri Becquerel", "Marie Curie", "Pierre Curie"], "CHEMICAL_ELEMENT": ["uranium", "thorium", "polonium", "radium"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION, CHEMICAL_COMPOUND, SCIENTIST, CHEMICAL_ELEMENT\n\nText:\nRadioactivity was discovered in 1896 by Henri Becquerel in uranium, and subsequently observed by Marie Curie and Pierre Curie in thorium and in the new elements polonium and radium.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION, CHEMICAL_COMPOUND, SCIENTIST, CHEMICAL_ELEMENT\n\nText:\nRadioactivity was discovered in 1896 by Henri Becquerel in uranium, and subsequently observed by Marie Curie and Pierre Curie in thorium and in the new elements polonium and radium.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_440", "text": "Investigations conducted at the site by William Duncan Strong, Waldo Rudolph Wedel, and A. T. Hill were instrumental in the development of Great Plains archaeology, and of Pawnee archaeology in particular.", "requested_types": ["ASTRONOMICAL_OBJECT", "MISC", "ORGANISATION", "SCIENTIST"], "distractors": ["ORGANISATION", "ASTRONOMICAL_OBJECT"], "gold": {"ASTRONOMICAL_OBJECT": [], "MISC": ["Great Plains archaeology", "Pawnee archaeology"], "ORGANISATION": [], "SCIENTIST": ["William Duncan Strong", "Waldo Rudolph Wedel", "A. T. Hill"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, ORGANISATION, SCIENTIST\n\nText:\nInvestigations conducted at the site by William Duncan Strong, Waldo Rudolph Wedel, and A. T. Hill were instrumental in the development of Great Plains archaeology, and of Pawnee archaeology in particular.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, ORGANISATION, SCIENTIST\n\nText:\nInvestigations conducted at the site by William Duncan Strong, Waldo Rudolph Wedel, and A. T. Hill were instrumental in the development of Great Plains archaeology, and of Pawnee archaeology in particular.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_441", "text": "In fact, when math E / math belongs to the spectrum, the inequality becomes an equality (the Aubry-André formula), proved by Jean Bourgain and Svetlana Jitomirskaya.", "requested_types": ["ACADEMIC_JOURNAL", "MISC", "CHEMICAL_ELEMENT", "THEORY", "SCIENTIST"], "distractors": ["MISC", "ACADEMIC_JOURNAL", "CHEMICAL_ELEMENT"], "gold": {"ACADEMIC_JOURNAL": [], "MISC": [], "CHEMICAL_ELEMENT": [], "THEORY": ["Aubry-André formula"], "SCIENTIST": ["Jean Bourgain", "Svetlana Jitomirskaya"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, MISC, CHEMICAL_ELEMENT, THEORY, SCIENTIST\n\nText:\nIn fact, when math E / math belongs to the spectrum, the inequality becomes an equality (the Aubry-André formula), proved by Jean Bourgain and Svetlana Jitomirskaya.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, MISC, CHEMICAL_ELEMENT, THEORY, SCIENTIST\n\nText:\nIn fact, when math E / math belongs to the spectrum, the inequality becomes an equality (the Aubry-André formula), proved by Jean Bourgain and Svetlana Jitomirskaya.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_442", "text": "Other studied asteroids included 15 Eunomia, 43 Ariadne, 44 Nysa, and 624 Hektor.", "requested_types": ["CHEMICAL_ELEMENT", "ASTRONOMICAL_OBJECT", "EVENT"], "distractors": ["EVENT", "CHEMICAL_ELEMENT"], "gold": {"CHEMICAL_ELEMENT": [], "ASTRONOMICAL_OBJECT": ["15 Eunomia", "43 Ariadne", "44 Nysa", "624 Hektor"], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, ASTRONOMICAL_OBJECT, EVENT\n\nText:\nOther studied asteroids included 15 Eunomia, 43 Ariadne, 44 Nysa, and 624 Hektor.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, ASTRONOMICAL_OBJECT, EVENT\n\nText:\nOther studied asteroids included 15 Eunomia, 43 Ariadne, 44 Nysa, and 624 Hektor.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_443", "text": "There are four community colleges in the valley; Los Angeles Valley College in Valley Glen, Los Angeles Mission College in Sylmar, Los Angeles Pierce College in Woodland Hills and Glendale Community College in the College Hills neighborhood of Glendale.", "requested_types": ["AWARD", "UNIVERSITY", "LOCATION"], "distractors": ["AWARD"], "gold": {"AWARD": [], "UNIVERSITY": ["Los Angeles Valley College", "Los Angeles Mission College", "Los Angeles Pierce College", "Glendale Community College"], "LOCATION": ["Valley Glen", "Sylmar", "Woodland Hills", "College Hills neighborhood of Glendale"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, UNIVERSITY, LOCATION\n\nText:\nThere are four community colleges in the valley; Los Angeles Valley College in Valley Glen, Los Angeles Mission College in Sylmar, Los Angeles Pierce College in Woodland Hills and Glendale Community College in the College Hills neighborhood of Glendale.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, UNIVERSITY, LOCATION\n\nText:\nThere are four community colleges in the valley; Los Angeles Valley College in Valley Glen, Los Angeles Mission College in Sylmar, Los Angeles Pierce College in Woodland Hills and Glendale Community College in the College Hills neighborhood of Glendale.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_444", "text": "He was elected President of the International Council for Science in 2008, to start his term in 2011.. International Council for Science.", "requested_types": ["ENZYME", "SCIENTIST", "ACADEMIC_JOURNAL", "ORGANISATION"], "distractors": ["SCIENTIST", "ACADEMIC_JOURNAL", "ENZYME"], "gold": {"ENZYME": [], "SCIENTIST": [], "ACADEMIC_JOURNAL": [], "ORGANISATION": ["International Council for Science"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, SCIENTIST, ACADEMIC_JOURNAL, ORGANISATION\n\nText:\nHe was elected President of the International Council for Science in 2008, to start his term in 2011.. International Council for Science.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, SCIENTIST, ACADEMIC_JOURNAL, ORGANISATION\n\nText:\nHe was elected President of the International Council for Science in 2008, to start his term in 2011.. International Council for Science.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_445", "text": "To occur, this lobopodia-based fibroblast migration needs Nesprin, Integrin, RhoA, ROCK and myosin II.", "requested_types": ["PROTEIN", "MISC", "EVENT"], "distractors": ["EVENT", "MISC"], "gold": {"PROTEIN": ["Nesprin", "Integrin", "RhoA", "ROCK", "myosin II"], "MISC": [], "EVENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, EVENT\n\nText:\nTo occur, this lobopodia-based fibroblast migration needs Nesprin, Integrin, RhoA, ROCK and myosin II.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, EVENT\n\nText:\nTo occur, this lobopodia-based fibroblast migration needs Nesprin, Integrin, RhoA, ROCK and myosin II.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_446", "text": "Martin Karplus, Michael Levitt and Arieh Warshel received the 2013 Nobel Prize in Chemistry for the development of multiscale models for complex chemical systems.", "requested_types": ["THEORY", "AWARD", "SCIENTIST", "CHEMICAL_ELEMENT"], "distractors": ["THEORY", "CHEMICAL_ELEMENT"], "gold": {"THEORY": [], "AWARD": ["Nobel Prize in Chemistry"], "SCIENTIST": ["Martin Karplus", "Michael Levitt", "Arieh Warshel"], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, AWARD, SCIENTIST, CHEMICAL_ELEMENT\n\nText:\nMartin Karplus, Michael Levitt and Arieh Warshel received the 2013 Nobel Prize in Chemistry for the development of multiscale models for complex chemical systems.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, AWARD, SCIENTIST, CHEMICAL_ELEMENT\n\nText:\nMartin Karplus, Michael Levitt and Arieh Warshel received the 2013 Nobel Prize in Chemistry for the development of multiscale models for complex chemical systems.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_447", "text": "The group included the physicists Walther Bothe, Robert Döpel, Hans Geiger, Wolfgang Gentner, Wilhelm Hanle, Gerhard Hoffmann, and Joos.", "requested_types": ["SCIENTIST", "THEORY"], "distractors": ["THEORY"], "gold": {"SCIENTIST": ["Walther Bothe", "Robert Döpel", "Hans Geiger", "Wolfgang Gentner", "Wilhelm Hanle", "Gerhard Hoffmann", "Joos"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, THEORY\n\nText:\nThe group included the physicists Walther Bothe, Robert Döpel, Hans Geiger, Wolfgang Gentner, Wilhelm Hanle, Gerhard Hoffmann, and Joos.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, THEORY\n\nText:\nThe group included the physicists Walther Bothe, Robert Döpel, Hans Geiger, Wolfgang Gentner, Wilhelm Hanle, Gerhard Hoffmann, and Joos.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_448", "text": "In 1991, he obtained a Doctor of Science degree in Physics and Mathematics from Moscow State University and moved to Tokyo (Japan) in 1993 to teach astronomy in Hitotsubashi University.", "requested_types": ["COUNTRY", "LOCATION", "UNIVERSITY", "ACADEMIC_JOURNAL", "MISC", "DISCIPLINE", "CHEMICAL_ELEMENT"], "distractors": ["ACADEMIC_JOURNAL", "CHEMICAL_ELEMENT"], "gold": {"COUNTRY": ["Japan"], "LOCATION": ["Tokyo"], "UNIVERSITY": ["Moscow State University", "Hitotsubashi University"], "ACADEMIC_JOURNAL": [], "MISC": ["Doctor of Science"], "DISCIPLINE": ["Physics", "Mathematics", "astronomy"], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, LOCATION, UNIVERSITY, ACADEMIC_JOURNAL, MISC, DISCIPLINE, CHEMICAL_ELEMENT\n\nText:\nIn 1991, he obtained a Doctor of Science degree in Physics and Mathematics from Moscow State University and moved to Tokyo (Japan) in 1993 to teach astronomy in Hitotsubashi University.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, LOCATION, UNIVERSITY, ACADEMIC_JOURNAL, MISC, DISCIPLINE, CHEMICAL_ELEMENT\n\nText:\nIn 1991, he obtained a Doctor of Science degree in Physics and Mathematics from Moscow State University and moved to Tokyo (Japan) in 1993 to teach astronomy in Hitotsubashi University.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_449", "text": "For his discovery of the neutron, Chadwick was awarded the Hughes Medal by the Royal Society in 1932, the Nobel Prize in Physics in 1935, the Copley Medal in 1950 and the Franklin Medal in 1951.", "requested_types": ["ASTRONOMICAL_OBJECT", "SCIENTIST", "AWARD", "ORGANISATION"], "distractors": ["ASTRONOMICAL_OBJECT"], "gold": {"ASTRONOMICAL_OBJECT": [], "SCIENTIST": ["Chadwick"], "AWARD": ["Hughes Medal", "Nobel Prize in Physics", "Copley Medal", "Franklin Medal"], "ORGANISATION": ["Royal Society"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, SCIENTIST, AWARD, ORGANISATION\n\nText:\nFor his discovery of the neutron, Chadwick was awarded the Hughes Medal by the Royal Society in 1932, the Nobel Prize in Physics in 1935, the Copley Medal in 1950 and the Franklin Medal in 1951.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, SCIENTIST, AWARD, ORGANISATION\n\nText:\nFor his discovery of the neutron, Chadwick was awarded the Hughes Medal by the Royal Society in 1932, the Nobel Prize in Physics in 1935, the Copley Medal in 1950 and the Franklin Medal in 1951.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_450", "text": "Wilson won gold medals in the 800 metres at both the 2011 World Youth Championships in Athletics and the 2012 World Junior Championships in Athletics.", "requested_types": ["THEORY", "EVENT", "PERSON", "ASTRONOMICAL_OBJECT", "COUNTRY"], "distractors": ["ASTRONOMICAL_OBJECT", "THEORY", "COUNTRY"], "gold": {"THEORY": [], "EVENT": ["2011 World Youth Championships in Athletics", "2012 World Junior Championships in Athletics"], "PERSON": ["Wilson"], "ASTRONOMICAL_OBJECT": [], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, EVENT, PERSON, ASTRONOMICAL_OBJECT, COUNTRY\n\nText:\nWilson won gold medals in the 800 metres at both the 2011 World Youth Championships in Athletics and the 2012 World Junior Championships in Athletics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, EVENT, PERSON, ASTRONOMICAL_OBJECT, COUNTRY\n\nText:\nWilson won gold medals in the 800 metres at both the 2011 World Youth Championships in Athletics and the 2012 World Junior Championships in Athletics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_451", "text": "On February 12, 2005, he was given a Grammy Lifetime Achievement Award by The Recording Academy.", "requested_types": ["PROTEIN", "MISC", "ORGANISATION", "EVENT", "AWARD"], "distractors": ["EVENT", "PROTEIN", "MISC"], "gold": {"PROTEIN": [], "MISC": [], "ORGANISATION": ["The Recording Academy"], "EVENT": [], "AWARD": ["Grammy Lifetime Achievement Award"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, ORGANISATION, EVENT, AWARD\n\nText:\nOn February 12, 2005, he was given a Grammy Lifetime Achievement Award by The Recording Academy.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, ORGANISATION, EVENT, AWARD\n\nText:\nOn February 12, 2005, he was given a Grammy Lifetime Achievement Award by The Recording Academy.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_452", "text": "Chicago's cultural heavyweights, such as the Art Institute of Chicago, the Goodman Theatre, the Chicago Theatre, the Lyric Opera at the Civic Opera House building, and the Chicago Symphony Orchestra, are also in this area, as is the historic Palmer House Hilton hotel, found on East Monroe Street.", "requested_types": ["EVENT", "ORGANISATION", "LOCATION", "ENZYME"], "distractors": ["EVENT", "ENZYME"], "gold": {"EVENT": [], "ORGANISATION": ["Art Institute of Chicago"], "LOCATION": ["Chicago", "Goodman Theatre", "Chicago Theatre", "Lyric Opera", "Civic Opera House", "Chicago Symphony Orchestra", "Palmer House Hilton hotel", "East Monroe Street"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, ORGANISATION, LOCATION, ENZYME\n\nText:\nChicago's cultural heavyweights, such as the Art Institute of Chicago, the Goodman Theatre, the Chicago Theatre, the Lyric Opera at the Civic Opera House building, and the Chicago Symphony Orchestra, are also in this area, as is the historic Palmer House Hilton hotel, found on East Monroe Street.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, ORGANISATION, LOCATION, ENZYME\n\nText:\nChicago's cultural heavyweights, such as the Art Institute of Chicago, the Goodman Theatre, the Chicago Theatre, the Lyric Opera at the Civic Opera House building, and the Chicago Symphony Orchestra, are also in this area, as is the historic Palmer House Hilton hotel, found on East Monroe Street.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_453", "text": "On 23 January 1960, the Swiss-designed Trieste, originally built in Italy and acquired by the U.S. Navy,, supported by the USS Wandank (ATF 204) and escorted by the USS Lewis (DE 535), descended to the ocean floor in the trench manned by Jacques Piccard (who co-designed the submersible along with his father, Auguste Piccard) and USN Lieutenant Don Walsh.", "requested_types": ["LOCATION", "MISC", "ORGANISATION", "SCIENTIST", "CHEMICAL_COMPOUND", "AWARD", "DISCIPLINE", "COUNTRY"], "distractors": ["DISCIPLINE", "AWARD", "CHEMICAL_COMPOUND"], "gold": {"LOCATION": ["Trieste"], "MISC": ["USS Wandank", "ATF 204", "USS Lewis"], "ORGANISATION": ["U.S. Navy"], "SCIENTIST": ["Jacques Piccard", "Auguste Piccard", "Don Walsh"], "CHEMICAL_COMPOUND": [], "AWARD": [], "DISCIPLINE": [], "COUNTRY": ["Italy"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, MISC, ORGANISATION, SCIENTIST, CHEMICAL_COMPOUND, AWARD, DISCIPLINE, COUNTRY\n\nText:\nOn 23 January 1960, the Swiss-designed Trieste, originally built in Italy and acquired by the U.S. Navy,, supported by the USS Wandank (ATF 204) and escorted by the USS Lewis (DE 535), descended to the ocean floor in the trench manned by Jacques Piccard (who co-designed the submersible along with his father, Auguste Piccard) and USN Lieutenant Don Walsh.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, MISC, ORGANISATION, SCIENTIST, CHEMICAL_COMPOUND, AWARD, DISCIPLINE, COUNTRY\n\nText:\nOn 23 January 1960, the Swiss-designed Trieste, originally built in Italy and acquired by the U.S. Navy,, supported by the USS Wandank (ATF 204) and escorted by the USS Lewis (DE 535), descended to the ocean floor in the trench manned by Jacques Piccard (who co-designed the submersible along with his father, Auguste Piccard) and USN Lieutenant Don Walsh.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_454", "text": "He has also received honorary doctorates from the universities of Newcastle, Surrey, Tel Aviv University, McGill, Simon Fraser University and the University of Tromsø.", "requested_types": ["ORGANISATION", "LOCATION", "UNIVERSITY", "MISC", "PERSON"], "distractors": ["ORGANISATION", "MISC", "PERSON"], "gold": {"ORGANISATION": [], "LOCATION": ["Surrey", "McGill"], "UNIVERSITY": ["universities of Newcastle", "Tel Aviv University", "Simon Fraser University", "University of Tromsø"], "MISC": [], "PERSON": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, LOCATION, UNIVERSITY, MISC, PERSON\n\nText:\nHe has also received honorary doctorates from the universities of Newcastle, Surrey, Tel Aviv University, McGill, Simon Fraser University and the University of Tromsø.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, LOCATION, UNIVERSITY, MISC, PERSON\n\nText:\nHe has also received honorary doctorates from the universities of Newcastle, Surrey, Tel Aviv University, McGill, Simon Fraser University and the University of Tromsø.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_455", "text": "Some fifty papers published in books and journals such as Nature, Philosophical Transactions of the Royal Society, Journal of Theoretical Biology, Journal of Zoology London, Journal of Animal Ecology, Animal Behaviour, and Proceedings of the Royal Society.", "requested_types": ["CHEMICAL_COMPOUND", "ACADEMIC_JOURNAL"], "distractors": ["CHEMICAL_COMPOUND"], "gold": {"CHEMICAL_COMPOUND": [], "ACADEMIC_JOURNAL": ["Nature", "Philosophical Transactions of the Royal Society", "Journal of Theoretical Biology", "Journal of Zoology London", "Journal of Animal Ecology", "Animal Behaviour", "Proceedings of the Royal Society"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ACADEMIC_JOURNAL\n\nText:\nSome fifty papers published in books and journals such as Nature, Philosophical Transactions of the Royal Society, Journal of Theoretical Biology, Journal of Zoology London, Journal of Animal Ecology, Animal Behaviour, and Proceedings of the Royal Society.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ACADEMIC_JOURNAL\n\nText:\nSome fifty papers published in books and journals such as Nature, Philosophical Transactions of the Royal Society, Journal of Theoretical Biology, Journal of Zoology London, Journal of Animal Ecology, Animal Behaviour, and Proceedings of the Royal Society.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_456", "text": "A conjunction of Venus and Jupiter occurred on 1 December 2008, and several hours later both planets separately reached conjunction with the crescent Moon.", "requested_types": ["ASTRONOMICAL_OBJECT", "PROTEIN", "COUNTRY"], "distractors": ["COUNTRY", "PROTEIN"], "gold": {"ASTRONOMICAL_OBJECT": ["Venus", "Jupiter", "Moon"], "PROTEIN": [], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, PROTEIN, COUNTRY\n\nText:\nA conjunction of Venus and Jupiter occurred on 1 December 2008, and several hours later both planets separately reached conjunction with the crescent Moon.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, PROTEIN, COUNTRY\n\nText:\nA conjunction of Venus and Jupiter occurred on 1 December 2008, and several hours later both planets separately reached conjunction with the crescent Moon.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_457", "text": "Repeated cocaine administration in mice induces post-translational modifications including hyperacetylation of Histone H3 or Histone H4 at 1,696 genes in one brain reward region the nucleus accumbens and deacetylation at 206 genes.", "requested_types": ["CHEMICAL_COMPOUND", "ORGANISATION", "MISC", "EVENT", "PROTEIN"], "distractors": ["EVENT", "ORGANISATION"], "gold": {"CHEMICAL_COMPOUND": ["cocaine"], "ORGANISATION": [], "MISC": ["post-translational modifications", "hyperacetylation", "nucleus accumbens", "deacetylation"], "EVENT": [], "PROTEIN": ["Histone H3", "Histone H4"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ORGANISATION, MISC, EVENT, PROTEIN\n\nText:\nRepeated cocaine administration in mice induces post-translational modifications including hyperacetylation of Histone H3 or Histone H4 at 1,696 genes in one brain reward region the nucleus accumbens and deacetylation at 206 genes.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, ORGANISATION, MISC, EVENT, PROTEIN\n\nText:\nRepeated cocaine administration in mice induces post-translational modifications including hyperacetylation of Histone H3 or Histone H4 at 1,696 genes in one brain reward region the nucleus accumbens and deacetylation at 206 genes.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_458", "text": "Others, such as Julian Steward and Leslie White, focused on how societies evolve and fit their ecological niche - an approach popularized by Marvin Harris.", "requested_types": ["THEORY", "SCIENTIST", "AWARD"], "distractors": ["THEORY", "AWARD"], "gold": {"THEORY": [], "SCIENTIST": ["Julian Steward", "Leslie White", "Marvin Harris"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, SCIENTIST, AWARD\n\nText:\nOthers, such as Julian Steward and Leslie White, focused on how societies evolve and fit their ecological niche - an approach popularized by Marvin Harris.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, SCIENTIST, AWARD\n\nText:\nOthers, such as Julian Steward and Leslie White, focused on how societies evolve and fit their ecological niche - an approach popularized by Marvin Harris.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_459", "text": "The invitees included Walther Bothe, Siegfried Flügge, Hans Geiger, Otto Hahn, Paul Harteck, Gerhard Hoffmann, Josef Mattauch, and Georg Stetter.", "requested_types": ["SCIENTIST", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND"], "gold": {"SCIENTIST": ["Walther Bothe", "Siegfried Flügge", "Hans Geiger", "Otto Hahn", "Paul Harteck", "Gerhard Hoffmann", "Josef Mattauch", "Georg Stetter"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_COMPOUND\n\nText:\nThe invitees included Walther Bothe, Siegfried Flügge, Hans Geiger, Otto Hahn, Paul Harteck, Gerhard Hoffmann, Josef Mattauch, and Georg Stetter.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_COMPOUND\n\nText:\nThe invitees included Walther Bothe, Siegfried Flügge, Hans Geiger, Otto Hahn, Paul Harteck, Gerhard Hoffmann, Josef Mattauch, and Georg Stetter.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_460", "text": "San Juan de Dios Hospital, Philippine Children's Medical Center, San Lazaro Hospital, Ospital ng Muntinlupa, Lung Center of the Philippines, the National Kidney and Transplant Institute, Manila Adventist Medical Center and Las Piñas Hospital also made steps to ban the toxic chemical.", "requested_types": ["UNIVERSITY", "MISC", "ORGANISATION", "PROTEIN"], "distractors": ["UNIVERSITY", "MISC", "PROTEIN"], "gold": {"UNIVERSITY": [], "MISC": [], "ORGANISATION": ["San Juan de Dios Hospital", "Philippine Children's Medical Center", "San Lazaro Hospital", "Ospital ng Muntinlupa", "Lung Center of the Philippines", "National Kidney and Transplant Institute", "Manila Adventist Medical Center", "Las Piñas Hospital"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, MISC, ORGANISATION, PROTEIN\n\nText:\nSan Juan de Dios Hospital, Philippine Children's Medical Center, San Lazaro Hospital, Ospital ng Muntinlupa, Lung Center of the Philippines, the National Kidney and Transplant Institute, Manila Adventist Medical Center and Las Piñas Hospital also made steps to ban the toxic chemical.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, MISC, ORGANISATION, PROTEIN\n\nText:\nSan Juan de Dios Hospital, Philippine Children's Medical Center, San Lazaro Hospital, Ospital ng Muntinlupa, Lung Center of the Philippines, the National Kidney and Transplant Institute, Manila Adventist Medical Center and Las Piñas Hospital also made steps to ban the toxic chemical.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_461", "text": "A similar example is given by the Senegalese sole (Solea senegalensis), which, when acclimated to temperatures of 26 ° C, produced a significantly higher amount of taurine, Glutamic acid, Gamma-Aminobutyric acid and glycine compared to acclimation to 12 ° C. This may mean that the aforementioned compounds aid in antioxidant defense, osmoregulatory processes, or energetic purposes at these temperatures.", "requested_types": ["CHEMICAL_COMPOUND", "MISC", "THEORY", "LOCATION", "PERSON"], "distractors": ["LOCATION", "PERSON"], "gold": {"CHEMICAL_COMPOUND": ["taurine", "Glutamic acid", "Gamma-Aminobutyric acid", "glycine"], "MISC": ["Senegalese sole", "Solea senegalensis"], "THEORY": ["antioxidant defense"], "LOCATION": [], "PERSON": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, MISC, THEORY, LOCATION, PERSON\n\nText:\nA similar example is given by the Senegalese sole (Solea senegalensis), which, when acclimated to temperatures of 26 ° C, produced a significantly higher amount of taurine, Glutamic acid, Gamma-Aminobutyric acid and glycine compared to acclimation to 12 ° C. This may mean that the aforementioned compounds aid in antioxidant defense, osmoregulatory processes, or energetic purposes at these temperatures.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, MISC, THEORY, LOCATION, PERSON\n\nText:\nA similar example is given by the Senegalese sole (Solea senegalensis), which, when acclimated to temperatures of 26 ° C, produced a significantly higher amount of taurine, Glutamic acid, Gamma-Aminobutyric acid and glycine compared to acclimation to 12 ° C. This may mean that the aforementioned compounds aid in antioxidant defense, osmoregulatory processes, or energetic purposes at these temperatures.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_462", "text": "Richard Laurence Millington Synge Fellow of the Royal Society FRSE FRIC FRSC MRIA (Liverpool, 28 October 1914 - Norwich, 18 August 1994) was a British biochemist, and shared the 1952 Nobel Prize in Chemistry for the invention of partition chromatography with Archer Martin.", "requested_types": ["ORGANISATION", "LOCATION", "SCIENTIST", "ENZYME", "AWARD", "PROTEIN", "MISC"], "distractors": ["ENZYME", "PROTEIN"], "gold": {"ORGANISATION": ["FRSE", "FRIC", "FRSC", "MRIA"], "LOCATION": ["Liverpool", "Norwich"], "SCIENTIST": ["Richard Laurence Millington Synge", "Archer Martin"], "ENZYME": [], "AWARD": ["Fellow of the Royal Society", "Nobel Prize in Chemistry"], "PROTEIN": [], "MISC": ["British", "partition chromatography"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, LOCATION, SCIENTIST, ENZYME, AWARD, PROTEIN, MISC\n\nText:\nRichard Laurence Millington Synge Fellow of the Royal Society FRSE FRIC FRSC MRIA (Liverpool, 28 October 1914 - Norwich, 18 August 1994) was a British biochemist, and shared the 1952 Nobel Prize in Chemistry for the invention of partition chromatography with Archer Martin.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, LOCATION, SCIENTIST, ENZYME, AWARD, PROTEIN, MISC\n\nText:\nRichard Laurence Millington Synge Fellow of the Royal Society FRSE FRIC FRSC MRIA (Liverpool, 28 October 1914 - Norwich, 18 August 1994) was a British biochemist, and shared the 1952 Nobel Prize in Chemistry for the invention of partition chromatography with Archer Martin.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_463", "text": "In December 2015, the Joint Working Party of international scientific bodies International Union of Pure and Applied Chemistry (IUPAC) and International Union of Pure and Applied Physics (IUPAP) recognized the element's discovery and assigned the priority of the discovery to the Dubna-Livermore collaboration..", "requested_types": ["LOCATION", "SCIENTIST", "ORGANISATION"], "distractors": ["LOCATION"], "gold": {"LOCATION": [], "SCIENTIST": ["Dubna-Livermore"], "ORGANISATION": ["Joint Working Party", "International Union of Pure and Applied Chemistry", "IUPAC", "International Union of Pure and Applied Physics", "IUPAP"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, SCIENTIST, ORGANISATION\n\nText:\nIn December 2015, the Joint Working Party of international scientific bodies International Union of Pure and Applied Chemistry (IUPAC) and International Union of Pure and Applied Physics (IUPAP) recognized the element's discovery and assigned the priority of the discovery to the Dubna-Livermore collaboration..\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, SCIENTIST, ORGANISATION\n\nText:\nIn December 2015, the Joint Working Party of international scientific bodies International Union of Pure and Applied Chemistry (IUPAC) and International Union of Pure and Applied Physics (IUPAP) recognized the element's discovery and assigned the priority of the discovery to the Dubna-Livermore collaboration..\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_464", "text": "He then studied meteorology and geography in Humboldt University of Berlin, where he had as instructors Ferdinand von Richthofen and Wilhelm von Bezold.", "requested_types": ["PERSON", "ENZYME", "SCIENTIST", "UNIVERSITY", "DISCIPLINE", "ASTRONOMICAL_OBJECT"], "distractors": ["PERSON", "ENZYME", "ASTRONOMICAL_OBJECT"], "gold": {"PERSON": [], "ENZYME": [], "SCIENTIST": ["Ferdinand von Richthofen", "Wilhelm von Bezold"], "UNIVERSITY": ["Humboldt University of Berlin"], "DISCIPLINE": ["meteorology", "geography"], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ENZYME, SCIENTIST, UNIVERSITY, DISCIPLINE, ASTRONOMICAL_OBJECT\n\nText:\nHe then studied meteorology and geography in Humboldt University of Berlin, where he had as instructors Ferdinand von Richthofen and Wilhelm von Bezold.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ENZYME, SCIENTIST, UNIVERSITY, DISCIPLINE, ASTRONOMICAL_OBJECT\n\nText:\nHe then studied meteorology and geography in Humboldt University of Berlin, where he had as instructors Ferdinand von Richthofen and Wilhelm von Bezold.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_465", "text": "Minimum distances from the Earth, Venus, and Jupiter, are 0.5, 0.8, and 3.5 AU, respectively.", "requested_types": ["CHEMICAL_COMPOUND", "DISCIPLINE", "PERSON", "ASTRONOMICAL_OBJECT"], "distractors": ["PERSON", "CHEMICAL_COMPOUND", "DISCIPLINE"], "gold": {"CHEMICAL_COMPOUND": [], "DISCIPLINE": [], "PERSON": [], "ASTRONOMICAL_OBJECT": ["Earth", "Venus", "Jupiter"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, DISCIPLINE, PERSON, ASTRONOMICAL_OBJECT\n\nText:\nMinimum distances from the Earth, Venus, and Jupiter, are 0.5, 0.8, and 3.5 AU, respectively.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, DISCIPLINE, PERSON, ASTRONOMICAL_OBJECT\n\nText:\nMinimum distances from the Earth, Venus, and Jupiter, are 0.5, 0.8, and 3.5 AU, respectively.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_466", "text": "Her published work includes biographies of Marie Curie, Alexander Graham Bell, Niels Bohr, Isaac Newton, Albert Einstein, Linus Pauling.", "requested_types": ["CHEMICAL_ELEMENT", "MISC", "SCIENTIST", "PROTEIN"], "distractors": ["MISC", "CHEMICAL_ELEMENT", "PROTEIN"], "gold": {"CHEMICAL_ELEMENT": [], "MISC": [], "SCIENTIST": ["Marie Curie", "Alexander Graham Bell", "Niels Bohr", "Isaac Newton", "Albert Einstein", "Linus Pauling"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, MISC, SCIENTIST, PROTEIN\n\nText:\nHer published work includes biographies of Marie Curie, Alexander Graham Bell, Niels Bohr, Isaac Newton, Albert Einstein, Linus Pauling.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, MISC, SCIENTIST, PROTEIN\n\nText:\nHer published work includes biographies of Marie Curie, Alexander Graham Bell, Niels Bohr, Isaac Newton, Albert Einstein, Linus Pauling.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_467", "text": "Friedmann Peak, in the Darwin Mountains of Antarctica, where she co-discovered endolith ic microorganisms in the Beacon sandstone, is named after her.", "requested_types": ["EVENT", "LOCATION", "COUNTRY"], "distractors": ["EVENT"], "gold": {"EVENT": [], "LOCATION": ["Friedmann Peak", "Darwin Mountains", "Beacon sandstone"], "COUNTRY": ["Antarctica"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, LOCATION, COUNTRY\n\nText:\nFriedmann Peak, in the Darwin Mountains of Antarctica, where she co-discovered endolith ic microorganisms in the Beacon sandstone, is named after her.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, LOCATION, COUNTRY\n\nText:\nFriedmann Peak, in the Darwin Mountains of Antarctica, where she co-discovered endolith ic microorganisms in the Beacon sandstone, is named after her.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_468", "text": "Three of the newly confirmed exoplanets were found to orbit within habitable zone s of their related star s: two of the three, Kepler-438b and Kepler-442b, are near-Earth-size and likely rocky; the third, Kepler-440b, is a super-Earth.", "requested_types": ["ORGANISATION", "MISC", "PERSON", "COUNTRY", "ASTRONOMICAL_OBJECT"], "distractors": ["ORGANISATION", "COUNTRY", "PERSON"], "gold": {"ORGANISATION": [], "MISC": ["exoplanets", "habitable zone", "super-Earth"], "PERSON": [], "COUNTRY": [], "ASTRONOMICAL_OBJECT": ["Kepler-438b", "Kepler-442b", "Kepler-440b"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, PERSON, COUNTRY, ASTRONOMICAL_OBJECT\n\nText:\nThree of the newly confirmed exoplanets were found to orbit within habitable zone s of their related star s: two of the three, Kepler-438b and Kepler-442b, are near-Earth-size and likely rocky; the third, Kepler-440b, is a super-Earth.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, PERSON, COUNTRY, ASTRONOMICAL_OBJECT\n\nText:\nThree of the newly confirmed exoplanets were found to orbit within habitable zone s of their related star s: two of the three, Kepler-438b and Kepler-442b, are near-Earth-size and likely rocky; the third, Kepler-440b, is a super-Earth.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_469", "text": "In the middle of the century, the Boone and Crockett Club and the Safari Club International developed complex scoring systems based on various dimensions and the number of tines or points, and they keep extensive records of high-scoring antlers.", "requested_types": ["DISCIPLINE", "ORGANISATION"], "distractors": ["DISCIPLINE"], "gold": {"DISCIPLINE": [], "ORGANISATION": ["Boone and Crockett Club", "Safari Club International"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ORGANISATION\n\nText:\nIn the middle of the century, the Boone and Crockett Club and the Safari Club International developed complex scoring systems based on various dimensions and the number of tines or points, and they keep extensive records of high-scoring antlers.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, ORGANISATION\n\nText:\nIn the middle of the century, the Boone and Crockett Club and the Safari Club International developed complex scoring systems based on various dimensions and the number of tines or points, and they keep extensive records of high-scoring antlers.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_470", "text": "In 1958, George Beadle and Edward Tatum received the Nobel Prize for work in fungi showing that one gene produces one enzyme.", "requested_types": ["SCIENTIST", "AWARD", "ORGANISATION", "THEORY"], "distractors": ["ORGANISATION"], "gold": {"SCIENTIST": ["George Beadle", "Edward Tatum"], "AWARD": ["Nobel Prize"], "ORGANISATION": [], "THEORY": ["one gene produces one enzyme"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, ORGANISATION, THEORY\n\nText:\nIn 1958, George Beadle and Edward Tatum received the Nobel Prize for work in fungi showing that one gene produces one enzyme.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, ORGANISATION, THEORY\n\nText:\nIn 1958, George Beadle and Edward Tatum received the Nobel Prize for work in fungi showing that one gene produces one enzyme.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_471", "text": "Oxidants are usually chemical substances with elements in high oxidation states, or else highly electronegative elements (Osub2 / sub, Fluorine, Chlorine, Bromine) that can gain extra electrons by oxidizing another substance.", "requested_types": ["CHEMICAL_ELEMENT", "UNIVERSITY", "CHEMICAL_COMPOUND"], "distractors": ["UNIVERSITY"], "gold": {"CHEMICAL_ELEMENT": ["Fluorine", "Chlorine", "Bromine"], "UNIVERSITY": [], "CHEMICAL_COMPOUND": ["Oxidants", "Osub2 / sub"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, UNIVERSITY, CHEMICAL_COMPOUND\n\nText:\nOxidants are usually chemical substances with elements in high oxidation states, or else highly electronegative elements (Osub2 / sub, Fluorine, Chlorine, Bromine) that can gain extra electrons by oxidizing another substance.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, UNIVERSITY, CHEMICAL_COMPOUND\n\nText:\nOxidants are usually chemical substances with elements in high oxidation states, or else highly electronegative elements (Osub2 / sub, Fluorine, Chlorine, Bromine) that can gain extra electrons by oxidizing another substance.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_472", "text": "While there, Bugs meets Marvin the Martian who is trying to blow up Earth with the Illudium Q-36 Explosive Space Modulator (in reality a stick of dynamite; this is the same device he tried to use in his debut short Haredevil Hare, though that Space Modulator was Uranium rather than Illudium) because it obstructs his view of Venus.", "requested_types": ["ASTRONOMICAL_OBJECT", "PERSON", "MISC", "AWARD"], "distractors": ["PERSON", "AWARD"], "gold": {"ASTRONOMICAL_OBJECT": ["Earth", "Uranium", "Venus"], "PERSON": [], "MISC": ["Marvin the Martian", "Illudium Q-36 Explosive Space Modulator", "Haredevil Hare", "Space Modulator"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, PERSON, MISC, AWARD\n\nText:\nWhile there, Bugs meets Marvin the Martian who is trying to blow up Earth with the Illudium Q-36 Explosive Space Modulator (in reality a stick of dynamite; this is the same device he tried to use in his debut short Haredevil Hare, though that Space Modulator was Uranium rather than Illudium) because it obstructs his view of Venus.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, PERSON, MISC, AWARD\n\nText:\nWhile there, Bugs meets Marvin the Martian who is trying to blow up Earth with the Illudium Q-36 Explosive Space Modulator (in reality a stick of dynamite; this is the same device he tried to use in his debut short Haredevil Hare, though that Space Modulator was Uranium rather than Illudium) because it obstructs his view of Venus.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_473", "text": "Their engagement with specific ideas relating to feminism, technoscience, political consciousness, and other social issues, formed the images and narrative of Haraway's book Modest _ Witness for which she received the Society for Social Studies of Science 's (4S) Ludwik Fleck Prize in 1999.", "requested_types": ["MISC", "THEORY", "AWARD", "SCIENTIST", "ORGANISATION"], "distractors": ["THEORY"], "gold": {"MISC": ["feminism", "technoscience", "political consciousness", "Modest _ Witness"], "THEORY": [], "AWARD": ["Ludwik Fleck Prize"], "SCIENTIST": ["Haraway"], "ORGANISATION": ["Society for Social Studies of Science", "4S"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, THEORY, AWARD, SCIENTIST, ORGANISATION\n\nText:\nTheir engagement with specific ideas relating to feminism, technoscience, political consciousness, and other social issues, formed the images and narrative of Haraway's book Modest _ Witness for which she received the Society for Social Studies of Science 's (4S) Ludwik Fleck Prize in 1999.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, THEORY, AWARD, SCIENTIST, ORGANISATION\n\nText:\nTheir engagement with specific ideas relating to feminism, technoscience, political consciousness, and other social issues, formed the images and narrative of Haraway's book Modest _ Witness for which she received the Society for Social Studies of Science 's (4S) Ludwik Fleck Prize in 1999.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_474", "text": "The next year, in 1995, a film adaptation of the book, Apollo 13, was released, directed by Ron Howard and starring Tom Hanks as Lovell, Bill Paxton as Haise, Kevin Bacon as Swigert, Gary Sinise as Mattingly, Ed Harris as Kranz, and Kathleen Quinlan as Marilyn Lovell.", "requested_types": ["EVENT", "LOCATION", "ORGANISATION", "MISC", "PERSON"], "distractors": ["ORGANISATION", "LOCATION", "EVENT"], "gold": {"EVENT": [], "LOCATION": [], "ORGANISATION": [], "MISC": ["Apollo 13"], "PERSON": ["Ron Howard", "Tom Hanks", "Lovell", "Bill Paxton", "Haise", "Kevin Bacon", "Swigert", "Gary Sinise", "Mattingly", "Ed Harris", "Kranz", "Kathleen Quinlan", "Marilyn Lovell"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, LOCATION, ORGANISATION, MISC, PERSON\n\nText:\nThe next year, in 1995, a film adaptation of the book, Apollo 13, was released, directed by Ron Howard and starring Tom Hanks as Lovell, Bill Paxton as Haise, Kevin Bacon as Swigert, Gary Sinise as Mattingly, Ed Harris as Kranz, and Kathleen Quinlan as Marilyn Lovell.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, LOCATION, ORGANISATION, MISC, PERSON\n\nText:\nThe next year, in 1995, a film adaptation of the book, Apollo 13, was released, directed by Ron Howard and starring Tom Hanks as Lovell, Bill Paxton as Haise, Kevin Bacon as Swigert, Gary Sinise as Mattingly, Ed Harris as Kranz, and Kathleen Quinlan as Marilyn Lovell.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_475", "text": "Only when a virulence factor is sensed, the Adenosine diphosphate is exchanged for Adenosine triphosphate.", "requested_types": ["CHEMICAL_COMPOUND", "MISC", "DISCIPLINE", "LOCATION"], "distractors": ["DISCIPLINE", "LOCATION"], "gold": {"CHEMICAL_COMPOUND": ["Adenosine diphosphate", "Adenosine triphosphate"], "MISC": ["virulence factor"], "DISCIPLINE": [], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, MISC, DISCIPLINE, LOCATION\n\nText:\nOnly when a virulence factor is sensed, the Adenosine diphosphate is exchanged for Adenosine triphosphate.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, MISC, DISCIPLINE, LOCATION\n\nText:\nOnly when a virulence factor is sensed, the Adenosine diphosphate is exchanged for Adenosine triphosphate.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_476", "text": "The awards that Mayr received include the National Medal of Science, the Balzan Prize, the Sarton Medal of the History of Science Society, the International Prize for Biology, the Loye and Alden Miller Research Award, and the Lewis Thomas Prize for Writing about Science.", "requested_types": ["SCIENTIST", "ENZYME", "AWARD", "CHEMICAL_COMPOUND"], "distractors": ["CHEMICAL_COMPOUND", "ENZYME"], "gold": {"SCIENTIST": ["Mayr"], "ENZYME": [], "AWARD": ["National Medal of Science", "Balzan Prize", "Sarton Medal of the History of Science Society", "International Prize for Biology", "Loye and Alden Miller Research Award", "Lewis Thomas Prize for Writing about Science"], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ENZYME, AWARD, CHEMICAL_COMPOUND\n\nText:\nThe awards that Mayr received include the National Medal of Science, the Balzan Prize, the Sarton Medal of the History of Science Society, the International Prize for Biology, the Loye and Alden Miller Research Award, and the Lewis Thomas Prize for Writing about Science.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ENZYME, AWARD, CHEMICAL_COMPOUND\n\nText:\nThe awards that Mayr received include the National Medal of Science, the Balzan Prize, the Sarton Medal of the History of Science Society, the International Prize for Biology, the Loye and Alden Miller Research Award, and the Lewis Thomas Prize for Writing about Science.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_477", "text": "In 1893 he received the Royal Society 's Copley Medal, then the most prestigious scientific prize in the world, for his researches and discoveries in physical science.", "requested_types": ["ORGANISATION", "DISCIPLINE", "AWARD", "ASTRONOMICAL_OBJECT"], "distractors": ["ASTRONOMICAL_OBJECT"], "gold": {"ORGANISATION": ["Royal Society"], "DISCIPLINE": ["physical science"], "AWARD": ["Copley Medal"], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, DISCIPLINE, AWARD, ASTRONOMICAL_OBJECT\n\nText:\nIn 1893 he received the Royal Society 's Copley Medal, then the most prestigious scientific prize in the world, for his researches and discoveries in physical science.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, DISCIPLINE, AWARD, ASTRONOMICAL_OBJECT\n\nText:\nIn 1893 he received the Royal Society 's Copley Medal, then the most prestigious scientific prize in the world, for his researches and discoveries in physical science.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_478", "text": "Santos's scientific work has appeared in journals such as Psychological Science, Animal Cognition, Developmental Science, Current Biology, Animal Behavior, and Cognition.", "requested_types": ["ACADEMIC_JOURNAL", "PERSON", "SCIENTIST"], "distractors": ["PERSON"], "gold": {"ACADEMIC_JOURNAL": ["Psychological Science", "Animal Cognition", "Developmental Science", "Current Biology", "Animal Behavior, and Cognition"], "PERSON": [], "SCIENTIST": ["Santos"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, PERSON, SCIENTIST\n\nText:\nSantos's scientific work has appeared in journals such as Psychological Science, Animal Cognition, Developmental Science, Current Biology, Animal Behavior, and Cognition.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, PERSON, SCIENTIST\n\nText:\nSantos's scientific work has appeared in journals such as Psychological Science, Animal Cognition, Developmental Science, Current Biology, Animal Behavior, and Cognition.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_479", "text": "The resultant increase in Aryl hydrocarbon receptor repressor expression could lead to a decrease in the body's ability to break down carcinogens, increasing the risk of cancer.", "requested_types": ["CHEMICAL_COMPOUND", "PROTEIN", "MISC"], "distractors": ["PROTEIN", "CHEMICAL_COMPOUND"], "gold": {"CHEMICAL_COMPOUND": [], "PROTEIN": [], "MISC": ["Aryl hydrocarbon receptor repressor", "carcinogens"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, PROTEIN, MISC\n\nText:\nThe resultant increase in Aryl hydrocarbon receptor repressor expression could lead to a decrease in the body's ability to break down carcinogens, increasing the risk of cancer.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, PROTEIN, MISC\n\nText:\nThe resultant increase in Aryl hydrocarbon receptor repressor expression could lead to a decrease in the body's ability to break down carcinogens, increasing the risk of cancer.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_480", "text": "His father encouraged him to apply to United States Military Academy, but he decided to enroll in the United States Naval Academy instead.", "requested_types": ["CHEMICAL_ELEMENT", "MISC", "UNIVERSITY", "LOCATION"], "distractors": ["LOCATION", "CHEMICAL_ELEMENT", "MISC"], "gold": {"CHEMICAL_ELEMENT": [], "MISC": [], "UNIVERSITY": ["United States Military Academy", "United States Naval Academy"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, MISC, UNIVERSITY, LOCATION\n\nText:\nHis father encouraged him to apply to United States Military Academy, but he decided to enroll in the United States Naval Academy instead.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_ELEMENT, MISC, UNIVERSITY, LOCATION\n\nText:\nHis father encouraged him to apply to United States Military Academy, but he decided to enroll in the United States Naval Academy instead.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_481", "text": "This would be achieved using biotechnology methods to confer to them the capacity to alter nucleotides in the chromosomes of infected individuals through sequence-specific editing systems like CRISPR, ZFN or TALEN.", "requested_types": ["PERSON", "EVENT", "CHEMICAL_COMPOUND", "MISC", "DISCIPLINE"], "distractors": ["PERSON", "EVENT"], "gold": {"PERSON": [], "EVENT": [], "CHEMICAL_COMPOUND": ["nucleotides"], "MISC": ["chromosomes", "CRISPR", "ZFN", "TALEN"], "DISCIPLINE": ["biotechnology"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, EVENT, CHEMICAL_COMPOUND, MISC, DISCIPLINE\n\nText:\nThis would be achieved using biotechnology methods to confer to them the capacity to alter nucleotides in the chromosomes of infected individuals through sequence-specific editing systems like CRISPR, ZFN or TALEN.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, EVENT, CHEMICAL_COMPOUND, MISC, DISCIPLINE\n\nText:\nThis would be achieved using biotechnology methods to confer to them the capacity to alter nucleotides in the chromosomes of infected individuals through sequence-specific editing systems like CRISPR, ZFN or TALEN.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_482", "text": "Toutatis is also a Mars-crosser asteroid with a chaotic orbit produced by a 3: 1 resonance with the planet Jupiter, a 1: 4 resonance with the planet Earth, and frequent close approaches to the terrestrial planet s, including Earth.", "requested_types": ["ASTRONOMICAL_OBJECT", "MISC", "DISCIPLINE"], "distractors": ["DISCIPLINE"], "gold": {"ASTRONOMICAL_OBJECT": ["Toutatis", "Jupiter", "Earth"], "MISC": ["Mars-crosser asteroid", "terrestrial planet"], "DISCIPLINE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, DISCIPLINE\n\nText:\nToutatis is also a Mars-crosser asteroid with a chaotic orbit produced by a 3: 1 resonance with the planet Jupiter, a 1: 4 resonance with the planet Earth, and frequent close approaches to the terrestrial planet s, including Earth.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC, DISCIPLINE\n\nText:\nToutatis is also a Mars-crosser asteroid with a chaotic orbit produced by a 3: 1 resonance with the planet Jupiter, a 1: 4 resonance with the planet Earth, and frequent close approaches to the terrestrial planet s, including Earth.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_483", "text": "It was discovered on 22 February 1938, by Finnish astronomer Yrjö Väisälä at Turku Observatory in Southwest Finland, and later named for Johannes Gabriel Granö, rector of the University of Turku.", "requested_types": ["UNIVERSITY", "ORGANISATION", "PERSON", "MISC", "SCIENTIST", "LOCATION"], "distractors": ["ORGANISATION", "PERSON"], "gold": {"UNIVERSITY": ["University of Turku"], "ORGANISATION": [], "PERSON": [], "MISC": ["Finnish"], "SCIENTIST": ["Yrjö Väisälä", "Turku Observatory", "Johannes Gabriel Granö"], "LOCATION": ["Southwest Finland"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ORGANISATION, PERSON, MISC, SCIENTIST, LOCATION\n\nText:\nIt was discovered on 22 February 1938, by Finnish astronomer Yrjö Väisälä at Turku Observatory in Southwest Finland, and later named for Johannes Gabriel Granö, rector of the University of Turku.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, ORGANISATION, PERSON, MISC, SCIENTIST, LOCATION\n\nText:\nIt was discovered on 22 February 1938, by Finnish astronomer Yrjö Väisälä at Turku Observatory in Southwest Finland, and later named for Johannes Gabriel Granö, rector of the University of Turku.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_484", "text": "Her research group demonstrated that one such small protein, AcrZ, binds to the multidrug efflux pump protein Acriflavine resistance protein family to affect its ability to export certain classes of antibiotics.", "requested_types": ["PROTEIN", "CHEMICAL_ELEMENT", "UNIVERSITY", "MISC"], "distractors": ["CHEMICAL_ELEMENT", "UNIVERSITY"], "gold": {"PROTEIN": ["AcrZ"], "CHEMICAL_ELEMENT": [], "UNIVERSITY": [], "MISC": ["Acriflavine resistance protein family", "antibiotics"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_ELEMENT, UNIVERSITY, MISC\n\nText:\nHer research group demonstrated that one such small protein, AcrZ, binds to the multidrug efflux pump protein Acriflavine resistance protein family to affect its ability to export certain classes of antibiotics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, CHEMICAL_ELEMENT, UNIVERSITY, MISC\n\nText:\nHer research group demonstrated that one such small protein, AcrZ, binds to the multidrug efflux pump protein Acriflavine resistance protein family to affect its ability to export certain classes of antibiotics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_485", "text": "Starling's successor, Archibald Hill, fostered the career of Bernard Katz, whose long association with UCL began in 1935.", "requested_types": ["MISC", "UNIVERSITY", "SCIENTIST", "ASTRONOMICAL_OBJECT"], "distractors": ["MISC", "ASTRONOMICAL_OBJECT"], "gold": {"MISC": [], "UNIVERSITY": ["UCL"], "SCIENTIST": ["Starling", "Archibald Hill", "Bernard Katz"], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, UNIVERSITY, SCIENTIST, ASTRONOMICAL_OBJECT\n\nText:\nStarling's successor, Archibald Hill, fostered the career of Bernard Katz, whose long association with UCL began in 1935.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, UNIVERSITY, SCIENTIST, ASTRONOMICAL_OBJECT\n\nText:\nStarling's successor, Archibald Hill, fostered the career of Bernard Katz, whose long association with UCL began in 1935.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_486", "text": "It was discovered on 19 September 1973, by Dutch astronomers Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plates taken by Tom Gehrels at the Palomar Observatory in California, United States.", "requested_types": ["PROTEIN", "MISC", "CHEMICAL_ELEMENT", "LOCATION", "SCIENTIST", "COUNTRY", "AWARD"], "distractors": ["AWARD", "PROTEIN", "CHEMICAL_ELEMENT"], "gold": {"PROTEIN": [], "MISC": ["Dutch"], "CHEMICAL_ELEMENT": [], "LOCATION": ["Leiden", "Palomar Observatory", "California"], "SCIENTIST": ["Ingrid van Houten-Groeneveld", "Cornelis van Houten", "Tom Gehrels"], "COUNTRY": ["United States"], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, CHEMICAL_ELEMENT, LOCATION, SCIENTIST, COUNTRY, AWARD\n\nText:\nIt was discovered on 19 September 1973, by Dutch astronomers Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plates taken by Tom Gehrels at the Palomar Observatory in California, United States.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, CHEMICAL_ELEMENT, LOCATION, SCIENTIST, COUNTRY, AWARD\n\nText:\nIt was discovered on 19 September 1973, by Dutch astronomers Ingrid van Houten-Groeneveld and Cornelis van Houten at Leiden, on photographic plates taken by Tom Gehrels at the Palomar Observatory in California, United States.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_487", "text": "Though Mendeleev was widely honored by scientific organizations all over Europe, including (in 1882) the Davy Medal from the Royal Society of London (which later also awarded him the Copley Medal in 1905),", "requested_types": ["LOCATION", "PROTEIN", "AWARD", "THEORY", "DISCIPLINE", "SCIENTIST"], "distractors": ["DISCIPLINE", "PROTEIN", "THEORY"], "gold": {"LOCATION": ["Europe"], "PROTEIN": [], "AWARD": ["Davy Medal from the Royal Society of London", "Copley Medal"], "THEORY": [], "DISCIPLINE": [], "SCIENTIST": ["Mendeleev"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, PROTEIN, AWARD, THEORY, DISCIPLINE, SCIENTIST\n\nText:\nThough Mendeleev was widely honored by scientific organizations all over Europe, including (in 1882) the Davy Medal from the Royal Society of London (which later also awarded him the Copley Medal in 1905),\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, PROTEIN, AWARD, THEORY, DISCIPLINE, SCIENTIST\n\nText:\nThough Mendeleev was widely honored by scientific organizations all over Europe, including (in 1882) the Davy Medal from the Royal Society of London (which later also awarded him the Copley Medal in 1905),\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_488", "text": "Ras stimulates Mitogen-activated protein kinase to phosphorylate ERK 1 / 2 which induce outgrowth of neurite s.", "requested_types": ["ORGANISATION", "MISC", "ENZYME", "PROTEIN"], "distractors": ["ORGANISATION"], "gold": {"ORGANISATION": [], "MISC": ["neurite"], "ENZYME": ["Mitogen-activated protein kinase"], "PROTEIN": ["Ras", "phosphorylate ERK 1 / 2"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, ENZYME, PROTEIN\n\nText:\nRas stimulates Mitogen-activated protein kinase to phosphorylate ERK 1 / 2 which induce outgrowth of neurite s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, MISC, ENZYME, PROTEIN\n\nText:\nRas stimulates Mitogen-activated protein kinase to phosphorylate ERK 1 / 2 which induce outgrowth of neurite s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_489", "text": "The results of his preliminary investigations were published in the journal Physikalische Zeitschrift in 1916 (a full paper was later published in Zeitschrift für Physik in 1922).", "requested_types": ["ACADEMIC_JOURNAL", "COUNTRY", "SCIENTIST", "ENZYME"], "distractors": ["SCIENTIST", "COUNTRY", "ENZYME"], "gold": {"ACADEMIC_JOURNAL": ["Physikalische Zeitschrift", "Zeitschrift für Physik"], "COUNTRY": [], "SCIENTIST": [], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, COUNTRY, SCIENTIST, ENZYME\n\nText:\nThe results of his preliminary investigations were published in the journal Physikalische Zeitschrift in 1916 (a full paper was later published in Zeitschrift für Physik in 1922).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, COUNTRY, SCIENTIST, ENZYME\n\nText:\nThe results of his preliminary investigations were published in the journal Physikalische Zeitschrift in 1916 (a full paper was later published in Zeitschrift für Physik in 1922).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_490", "text": "Hund worked at the Universities of University of Rostock, Leipzig, University of Jena, Frankfurt am Main, and University of Göttingen.", "requested_types": ["LOCATION", "EVENT", "DISCIPLINE", "SCIENTIST", "UNIVERSITY"], "distractors": ["DISCIPLINE", "EVENT"], "gold": {"LOCATION": ["Leipzig", "Frankfurt am Main"], "EVENT": [], "DISCIPLINE": [], "SCIENTIST": ["Hund"], "UNIVERSITY": ["University of Rostock", "University of Jena", "University of Göttingen"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, EVENT, DISCIPLINE, SCIENTIST, UNIVERSITY\n\nText:\nHund worked at the Universities of University of Rostock, Leipzig, University of Jena, Frankfurt am Main, and University of Göttingen.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: LOCATION, EVENT, DISCIPLINE, SCIENTIST, UNIVERSITY\n\nText:\nHund worked at the Universities of University of Rostock, Leipzig, University of Jena, Frankfurt am Main, and University of Göttingen.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_491", "text": "He was awarded the William Bowie Medal by the American Geophysical Union, the Alexander Agassiz Medal of the National Academy of Sciences, the Medal of the Royal Geographical Society, the Vega Medal by the Swedish Society for Anthropology and Geography and the Sweden Order of the Polar Star.", "requested_types": ["SCIENTIST", "AWARD", "ENZYME"], "distractors": ["SCIENTIST", "ENZYME"], "gold": {"SCIENTIST": [], "AWARD": ["William Bowie Medal by the American Geophysical Union", "Alexander Agassiz Medal of the National Academy of Sciences", "Medal of the Royal Geographical Society", "Vega Medal by the Swedish Society for Anthropology and Geography", "Sweden Order of the Polar Star"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, ENZYME\n\nText:\nHe was awarded the William Bowie Medal by the American Geophysical Union, the Alexander Agassiz Medal of the National Academy of Sciences, the Medal of the Royal Geographical Society, the Vega Medal by the Swedish Society for Anthropology and Geography and the Sweden Order of the Polar Star.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, ENZYME\n\nText:\nHe was awarded the William Bowie Medal by the American Geophysical Union, the Alexander Agassiz Medal of the National Academy of Sciences, the Medal of the Royal Geographical Society, the Vega Medal by the Swedish Society for Anthropology and Geography and the Sweden Order of the Polar Star.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_492", "text": "For example, calcium antagonist s, which are used to treat coronary heart disease, hypertension, and cardiac arrhythmia, act by inhibiting L-type calcium channel or T-type calcium channel calcium channels.", "requested_types": ["EVENT", "CHEMICAL_ELEMENT", "DISCIPLINE", "CHEMICAL_COMPOUND", "MISC"], "distractors": ["EVENT", "CHEMICAL_COMPOUND", "DISCIPLINE"], "gold": {"EVENT": [], "CHEMICAL_ELEMENT": ["calcium"], "DISCIPLINE": [], "CHEMICAL_COMPOUND": [], "MISC": ["calcium antagonist", "coronary heart disease", "hypertension", "cardiac arrhythmia", "L-type calcium channel", "T-type calcium channel"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, CHEMICAL_ELEMENT, DISCIPLINE, CHEMICAL_COMPOUND, MISC\n\nText:\nFor example, calcium antagonist s, which are used to treat coronary heart disease, hypertension, and cardiac arrhythmia, act by inhibiting L-type calcium channel or T-type calcium channel calcium channels.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, CHEMICAL_ELEMENT, DISCIPLINE, CHEMICAL_COMPOUND, MISC\n\nText:\nFor example, calcium antagonist s, which are used to treat coronary heart disease, hypertension, and cardiac arrhythmia, act by inhibiting L-type calcium channel or T-type calcium channel calcium channels.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_493", "text": "The report on the B 8 experiment was written by Fritz Bopp, Erich Fischer, Werner Heisenberg, and Karl Wirtz from the KWIP and Walther Bothe, Peter Jensen, and Oskar Ritter from the KWImF.Hentschel and Hentschel, 1996, 377 and 377n73.", "requested_types": ["SCIENTIST", "ORGANISATION", "CHEMICAL_ELEMENT", "LOCATION"], "distractors": ["CHEMICAL_ELEMENT"], "gold": {"SCIENTIST": ["Fritz Bopp", "Erich Fischer", "Werner Heisenberg", "Karl Wirtz", "Walther Bothe", "Peter Jensen", "Oskar Ritter"], "ORGANISATION": ["KWIP", "KWImF.Hentschel"], "CHEMICAL_ELEMENT": [], "LOCATION": ["Hentschel"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ORGANISATION, CHEMICAL_ELEMENT, LOCATION\n\nText:\nThe report on the B 8 experiment was written by Fritz Bopp, Erich Fischer, Werner Heisenberg, and Karl Wirtz from the KWIP and Walther Bothe, Peter Jensen, and Oskar Ritter from the KWImF.Hentschel and Hentschel, 1996, 377 and 377n73.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, ORGANISATION, CHEMICAL_ELEMENT, LOCATION\n\nText:\nThe report on the B 8 experiment was written by Fritz Bopp, Erich Fischer, Werner Heisenberg, and Karl Wirtz from the KWIP and Walther Bothe, Peter Jensen, and Oskar Ritter from the KWImF.Hentschel and Hentschel, 1996, 377 and 377n73.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_494", "text": "In the 1960s, Ehlers collaborated with Felix Pirani and Alfred Schild on a constructive-axiomatic approach to general relativity: a way of deriving the theory from a minimal set of elementary objects and a set of axioms specifying these objects' properties.", "requested_types": ["UNIVERSITY", "THEORY", "SCIENTIST"], "distractors": ["UNIVERSITY"], "gold": {"UNIVERSITY": [], "THEORY": ["general relativity"], "SCIENTIST": ["Ehlers", "Felix Pirani", "Alfred Schild"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, THEORY, SCIENTIST\n\nText:\nIn the 1960s, Ehlers collaborated with Felix Pirani and Alfred Schild on a constructive-axiomatic approach to general relativity: a way of deriving the theory from a minimal set of elementary objects and a set of axioms specifying these objects' properties.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, THEORY, SCIENTIST\n\nText:\nIn the 1960s, Ehlers collaborated with Felix Pirani and Alfred Schild on a constructive-axiomatic approach to general relativity: a way of deriving the theory from a minimal set of elementary objects and a set of axioms specifying these objects' properties.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_495", "text": "Amphitrite (minor planet designation: 29 Amphridite) is one of the largest S-type asteroid s, approximately in diameter, and probably third largest after 15 Eunomia and 3 Juno, although 7 Iris and 532 Herculina are similar in size.", "requested_types": ["PROTEIN", "ASTRONOMICAL_OBJECT", "CHEMICAL_COMPOUND", "ORGANISATION", "MISC"], "distractors": ["CHEMICAL_COMPOUND", "ORGANISATION", "PROTEIN"], "gold": {"PROTEIN": [], "ASTRONOMICAL_OBJECT": ["Amphitrite", "29 Amphridite", "15 Eunomia", "3 Juno", "7 Iris", "532 Herculina"], "CHEMICAL_COMPOUND": [], "ORGANISATION": [], "MISC": ["minor planet"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, ORGANISATION, MISC\n\nText:\nAmphitrite (minor planet designation: 29 Amphridite) is one of the largest S-type asteroid s, approximately in diameter, and probably third largest after 15 Eunomia and 3 Juno, although 7 Iris and 532 Herculina are similar in size.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, ASTRONOMICAL_OBJECT, CHEMICAL_COMPOUND, ORGANISATION, MISC\n\nText:\nAmphitrite (minor planet designation: 29 Amphridite) is one of the largest S-type asteroid s, approximately in diameter, and probably third largest after 15 Eunomia and 3 Juno, although 7 Iris and 532 Herculina are similar in size.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_496", "text": "Loss of function MLH1, MSH2, MSH6 or PMS2 genes cause defective DNA mismatch repair.", "requested_types": ["ACADEMIC_JOURNAL", "MISC"], "distractors": ["ACADEMIC_JOURNAL"], "gold": {"ACADEMIC_JOURNAL": [], "MISC": ["MLH1", "MSH2", "MSH6", "PMS2", "DNA mismatch repair"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, MISC\n\nText:\nLoss of function MLH1, MSH2, MSH6 or PMS2 genes cause defective DNA mismatch repair.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, MISC\n\nText:\nLoss of function MLH1, MSH2, MSH6 or PMS2 genes cause defective DNA mismatch repair.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_497", "text": "In 2011, the chairman and ranking member of the United States Senate Committee on Armed Services called for an alternative plan where Futenma aircraft would move to Kadena Air Base while the current aircraft at Kadena would move to Andersen Air Force Base.", "requested_types": ["ORGANISATION", "THEORY", "LOCATION"], "distractors": ["THEORY"], "gold": {"ORGANISATION": ["United States Senate Committee on Armed Services"], "THEORY": [], "LOCATION": ["Futenma", "Kadena Air Base", "Kadena", "Andersen Air Force Base"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, THEORY, LOCATION\n\nText:\nIn 2011, the chairman and ranking member of the United States Senate Committee on Armed Services called for an alternative plan where Futenma aircraft would move to Kadena Air Base while the current aircraft at Kadena would move to Andersen Air Force Base.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, THEORY, LOCATION\n\nText:\nIn 2011, the chairman and ranking member of the United States Senate Committee on Armed Services called for an alternative plan where Futenma aircraft would move to Kadena Air Base while the current aircraft at Kadena would move to Andersen Air Force Base.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_498", "text": "Other members of the Nuclear Physics Working Group in both 1956 and 1957 were: Werner Heisenberg (chairman), Hans Kopfermann (vice-chairman), Fritz Bopp, Walther Bothe, Otto Haxel, Willibald Jentschke, Heinz Maier-Leibnitz, Josef Mattauch, Wolfgang Riezler, Wilhelm Walcher, and Carl Friedrich von Weizsäcker.", "requested_types": ["PERSON", "ORGANISATION", "UNIVERSITY", "SCIENTIST"], "distractors": ["UNIVERSITY"], "gold": {"PERSON": ["Willibald Jentschke"], "ORGANISATION": ["Nuclear Physics Working Group"], "UNIVERSITY": [], "SCIENTIST": ["Werner Heisenberg", "Hans Kopfermann", "Fritz Bopp", "Walther Bothe", "Otto Haxel", "Heinz Maier-Leibnitz", "Josef Mattauch", "Wolfgang Riezler", "Wilhelm Walcher", "Carl Friedrich von Weizsäcker"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION, UNIVERSITY, SCIENTIST\n\nText:\nOther members of the Nuclear Physics Working Group in both 1956 and 1957 were: Werner Heisenberg (chairman), Hans Kopfermann (vice-chairman), Fritz Bopp, Walther Bothe, Otto Haxel, Willibald Jentschke, Heinz Maier-Leibnitz, Josef Mattauch, Wolfgang Riezler, Wilhelm Walcher, and Carl Friedrich von Weizsäcker.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, ORGANISATION, UNIVERSITY, SCIENTIST\n\nText:\nOther members of the Nuclear Physics Working Group in both 1956 and 1957 were: Werner Heisenberg (chairman), Hans Kopfermann (vice-chairman), Fritz Bopp, Walther Bothe, Otto Haxel, Willibald Jentschke, Heinz Maier-Leibnitz, Josef Mattauch, Wolfgang Riezler, Wilhelm Walcher, and Carl Friedrich von Weizsäcker.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_499", "text": "There, he worked with scientists including Luis Walter Alvarez, John von Neumann and George Kistiakowsky on the development of the high explosives and exploding-bridgewire detonator s required by atomic bomb s.", "requested_types": ["SCIENTIST", "EVENT", "ORGANISATION", "MISC"], "distractors": ["ORGANISATION", "EVENT"], "gold": {"SCIENTIST": ["Luis Walter Alvarez", "John von Neumann", "George Kistiakowsky"], "EVENT": [], "ORGANISATION": [], "MISC": ["atomic bomb"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, EVENT, ORGANISATION, MISC\n\nText:\nThere, he worked with scientists including Luis Walter Alvarez, John von Neumann and George Kistiakowsky on the development of the high explosives and exploding-bridgewire detonator s required by atomic bomb s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, EVENT, ORGANISATION, MISC\n\nText:\nThere, he worked with scientists including Luis Walter Alvarez, John von Neumann and George Kistiakowsky on the development of the high explosives and exploding-bridgewire detonator s required by atomic bomb s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_500", "text": "He has collaborated with, among others, Luciano Pietronero, Benoit Mandelbrot, Betz Halloran, Ira Longini, and David Lazer.", "requested_types": ["UNIVERSITY", "THEORY", "SCIENTIST"], "distractors": ["UNIVERSITY", "THEORY"], "gold": {"UNIVERSITY": [], "THEORY": [], "SCIENTIST": ["Luciano Pietronero", "Benoit Mandelbrot", "Betz Halloran", "Ira Longini", "David Lazer"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, THEORY, SCIENTIST\n\nText:\nHe has collaborated with, among others, Luciano Pietronero, Benoit Mandelbrot, Betz Halloran, Ira Longini, and David Lazer.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, THEORY, SCIENTIST\n\nText:\nHe has collaborated with, among others, Luciano Pietronero, Benoit Mandelbrot, Betz Halloran, Ira Longini, and David Lazer.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_501", "text": "Several DNA markers, belonging to the nuclear, mitochondrial, and chloroplast genomes (RuBisCO, COI, ITS + 5.8S, SSU, 18S...), have been designed and successfully used for diatoms identification with NGS.", "requested_types": ["MISC", "ENZYME", "ORGANISATION"], "distractors": ["ORGANISATION"], "gold": {"MISC": ["DNA", "COI", "ITS", "5.8S", "SSU", "18S", "NGS"], "ENZYME": ["RuBisCO"], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ENZYME, ORGANISATION\n\nText:\nSeveral DNA markers, belonging to the nuclear, mitochondrial, and chloroplast genomes (RuBisCO, COI, ITS + 5.8S, SSU, 18S...), have been designed and successfully used for diatoms identification with NGS.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ENZYME, ORGANISATION\n\nText:\nSeveral DNA markers, belonging to the nuclear, mitochondrial, and chloroplast genomes (RuBisCO, COI, ITS + 5.8S, SSU, 18S...), have been designed and successfully used for diatoms identification with NGS.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_502", "text": "The Nobel Prize in Chemistry 1997 was divided, one half jointly to Paul D. Boyer and John E. Walker for their elucidation of the enzymatic mechanism underlying the synthesis of adenosine triphosphate (ATP) and the other half to Jens C. Skou for the first discovery of an ion-transporting enzyme, Na +, K + -ATPase.", "requested_types": ["SCIENTIST", "CHEMICAL_ELEMENT", "CHEMICAL_COMPOUND", "ENZYME", "PROTEIN", "AWARD"], "distractors": ["CHEMICAL_ELEMENT", "PROTEIN"], "gold": {"SCIENTIST": ["Paul D. Boyer", "John E. Walker", "Jens C. Skou"], "CHEMICAL_ELEMENT": [], "CHEMICAL_COMPOUND": ["adenosine triphosphate", "ATP"], "ENZYME": ["ion-transporting enzyme"], "PROTEIN": [], "AWARD": ["Nobel Prize in Chemistry"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_ELEMENT, CHEMICAL_COMPOUND, ENZYME, PROTEIN, AWARD\n\nText:\nThe Nobel Prize in Chemistry 1997 was divided, one half jointly to Paul D. Boyer and John E. Walker for their elucidation of the enzymatic mechanism underlying the synthesis of adenosine triphosphate (ATP) and the other half to Jens C. Skou for the first discovery of an ion-transporting enzyme, Na +, K + -ATPase.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, CHEMICAL_ELEMENT, CHEMICAL_COMPOUND, ENZYME, PROTEIN, AWARD\n\nText:\nThe Nobel Prize in Chemistry 1997 was divided, one half jointly to Paul D. Boyer and John E. Walker for their elucidation of the enzymatic mechanism underlying the synthesis of adenosine triphosphate (ATP) and the other half to Jens C. Skou for the first discovery of an ion-transporting enzyme, Na +, K + -ATPase.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_503", "text": "The 2013 Nobel Peace Prize was awarded to the organization because it had, with the Chemical Weapons Convention, defined the use of chemical weapons as a taboo under international law according to Thorbjørn Jagland, Chairman of the Norwegian Nobel Committee.", "requested_types": ["PERSON", "MISC", "ORGANISATION", "COUNTRY", "ACADEMIC_JOURNAL", "AWARD", "THEORY"], "distractors": ["THEORY", "ACADEMIC_JOURNAL", "COUNTRY"], "gold": {"PERSON": ["Thorbjørn Jagland"], "MISC": ["Chemical Weapons Convention"], "ORGANISATION": ["Norwegian Nobel Committee"], "COUNTRY": [], "ACADEMIC_JOURNAL": [], "AWARD": ["2013 Nobel Peace Prize"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, MISC, ORGANISATION, COUNTRY, ACADEMIC_JOURNAL, AWARD, THEORY\n\nText:\nThe 2013 Nobel Peace Prize was awarded to the organization because it had, with the Chemical Weapons Convention, defined the use of chemical weapons as a taboo under international law according to Thorbjørn Jagland, Chairman of the Norwegian Nobel Committee.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, MISC, ORGANISATION, COUNTRY, ACADEMIC_JOURNAL, AWARD, THEORY\n\nText:\nThe 2013 Nobel Peace Prize was awarded to the organization because it had, with the Chemical Weapons Convention, defined the use of chemical weapons as a taboo under international law according to Thorbjørn Jagland, Chairman of the Norwegian Nobel Committee.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_504", "text": "More recently, ecoimmunology has been the theme of three special issues in peer-reviewed journals, in Philosophical Transactions of the Royal Society B, in Functional Ecology, and in Physiological and Biochemical Zoology (see External links).", "requested_types": ["ACADEMIC_JOURNAL", "DISCIPLINE", "ORGANISATION", "SCIENTIST"], "distractors": ["SCIENTIST", "ORGANISATION", "DISCIPLINE"], "gold": {"ACADEMIC_JOURNAL": ["Philosophical Transactions of the Royal Society B", "Functional Ecology", "Physiological and Biochemical Zoology"], "DISCIPLINE": [], "ORGANISATION": [], "SCIENTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, DISCIPLINE, ORGANISATION, SCIENTIST\n\nText:\nMore recently, ecoimmunology has been the theme of three special issues in peer-reviewed journals, in Philosophical Transactions of the Royal Society B, in Functional Ecology, and in Physiological and Biochemical Zoology (see External links).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, DISCIPLINE, ORGANISATION, SCIENTIST\n\nText:\nMore recently, ecoimmunology has been the theme of three special issues in peer-reviewed journals, in Philosophical Transactions of the Royal Society B, in Functional Ecology, and in Physiological and Biochemical Zoology (see External links).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_505", "text": "Some notable examples include Journal of Computational Biology and PLOS Computational Biology.", "requested_types": ["ENZYME", "ACADEMIC_JOURNAL"], "distractors": ["ENZYME"], "gold": {"ENZYME": [], "ACADEMIC_JOURNAL": ["Journal of Computational Biology", "PLOS Computational Biology"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ACADEMIC_JOURNAL\n\nText:\nSome notable examples include Journal of Computational Biology and PLOS Computational Biology.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ENZYME, ACADEMIC_JOURNAL\n\nText:\nSome notable examples include Journal of Computational Biology and PLOS Computational Biology.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_506", "text": "Nectin s are a distinct family of cell adhesion molecules.", "requested_types": ["MISC", "ENZYME"], "distractors": ["ENZYME"], "gold": {"MISC": ["Nectin", "cell adhesion molecules"], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ENZYME\n\nText:\nNectin s are a distinct family of cell adhesion molecules.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ENZYME\n\nText:\nNectin s are a distinct family of cell adhesion molecules.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_507", "text": "He was a journal reviewer for Cell, Science, Nature, Nature Genetics, Nature Immunology, Nature Biotechnology, Proceedings of the National Academy of Sciences, Public Library of Science, Development, Genes Development, Journal of Clinical Investigation and Blood.", "requested_types": ["SCIENTIST", "AWARD", "ACADEMIC_JOURNAL"], "distractors": ["AWARD", "SCIENTIST"], "gold": {"SCIENTIST": [], "AWARD": [], "ACADEMIC_JOURNAL": ["Cell", "Science", "Nature", "Nature Genetics", "Nature Immunology", "Nature Biotechnology", "Proceedings of the National Academy of Sciences", "Public Library of Science, Development", "Genes Development", "Journal of Clinical Investigation", "Blood"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, ACADEMIC_JOURNAL\n\nText:\nHe was a journal reviewer for Cell, Science, Nature, Nature Genetics, Nature Immunology, Nature Biotechnology, Proceedings of the National Academy of Sciences, Public Library of Science, Development, Genes Development, Journal of Clinical Investigation and Blood.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, ACADEMIC_JOURNAL\n\nText:\nHe was a journal reviewer for Cell, Science, Nature, Nature Genetics, Nature Immunology, Nature Biotechnology, Proceedings of the National Academy of Sciences, Public Library of Science, Development, Genes Development, Journal of Clinical Investigation and Blood.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_508", "text": "Following their charting of the JNK signaling pathway, Karin and coworkers have begun to study the role of protein phosphorylation in control of NF-κB activity.", "requested_types": ["MISC", "PROTEIN", "DISCIPLINE", "SCIENTIST"], "distractors": ["DISCIPLINE"], "gold": {"MISC": ["JNK signaling pathway", "protein phosphorylation"], "PROTEIN": ["NF-κB"], "DISCIPLINE": [], "SCIENTIST": ["Karin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, DISCIPLINE, SCIENTIST\n\nText:\nFollowing their charting of the JNK signaling pathway, Karin and coworkers have begun to study the role of protein phosphorylation in control of NF-κB activity.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, PROTEIN, DISCIPLINE, SCIENTIST\n\nText:\nFollowing their charting of the JNK signaling pathway, Karin and coworkers have begun to study the role of protein phosphorylation in control of NF-κB activity.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_509", "text": "Two American explorers claimed to reach the North Pole; Frederick Cook in 1908 and Robert Peary in 1909.", "requested_types": ["MISC", "DISCIPLINE", "PERSON", "ENZYME", "LOCATION"], "distractors": ["ENZYME", "DISCIPLINE"], "gold": {"MISC": ["American"], "DISCIPLINE": [], "PERSON": ["Frederick Cook", "Robert Peary"], "ENZYME": [], "LOCATION": ["North Pole"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, DISCIPLINE, PERSON, ENZYME, LOCATION\n\nText:\nTwo American explorers claimed to reach the North Pole; Frederick Cook in 1908 and Robert Peary in 1909.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, DISCIPLINE, PERSON, ENZYME, LOCATION\n\nText:\nTwo American explorers claimed to reach the North Pole; Frederick Cook in 1908 and Robert Peary in 1909.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_510", "text": "A number of Pluto -as-planet proponents, in particular Alan Stern, head of NASA's New Horizons mission to Pluto, have circulated a petition among astronomers to alter the definition.", "requested_types": ["MISC", "ORGANISATION", "SCIENTIST", "ASTRONOMICAL_OBJECT", "THEORY"], "distractors": ["THEORY"], "gold": {"MISC": ["New Horizons"], "ORGANISATION": ["NASA"], "SCIENTIST": ["Alan Stern"], "ASTRONOMICAL_OBJECT": ["Pluto"], "THEORY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ORGANISATION, SCIENTIST, ASTRONOMICAL_OBJECT, THEORY\n\nText:\nA number of Pluto -as-planet proponents, in particular Alan Stern, head of NASA's New Horizons mission to Pluto, have circulated a petition among astronomers to alter the definition.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ORGANISATION, SCIENTIST, ASTRONOMICAL_OBJECT, THEORY\n\nText:\nA number of Pluto -as-planet proponents, in particular Alan Stern, head of NASA's New Horizons mission to Pluto, have circulated a petition among astronomers to alter the definition.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_511", "text": "In 1972, LaHaye helped establish the Institute for Creation Research at San Diego Christian College in El Cajon, California, along with Henry M. Morris.", "requested_types": ["ORGANISATION", "UNIVERSITY", "PERSON", "LOCATION", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"ORGANISATION": ["Institute for Creation Research"], "UNIVERSITY": ["San Diego Christian College"], "PERSON": ["LaHaye", "Henry M. Morris"], "LOCATION": ["El Cajon", "California"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, UNIVERSITY, PERSON, LOCATION, COUNTRY\n\nText:\nIn 1972, LaHaye helped establish the Institute for Creation Research at San Diego Christian College in El Cajon, California, along with Henry M. Morris.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ORGANISATION, UNIVERSITY, PERSON, LOCATION, COUNTRY\n\nText:\nIn 1972, LaHaye helped establish the Institute for Creation Research at San Diego Christian College in El Cajon, California, along with Henry M. Morris.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_512", "text": "Derek Abbott, Paul C.W. Davies, and Arun K. Pati, Quantum Aspects of Life, Imperial College Press, 2008, p.", "requested_types": ["SCIENTIST", "MISC", "DISCIPLINE", "ORGANISATION"], "distractors": ["DISCIPLINE"], "gold": {"SCIENTIST": ["Derek Abbott", "Paul C.W. Davies", "Arun K. Pati"], "MISC": ["Quantum Aspects of Life"], "DISCIPLINE": [], "ORGANISATION": ["Imperial College Press"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, MISC, DISCIPLINE, ORGANISATION\n\nText:\nDerek Abbott, Paul C.W. Davies, and Arun K. Pati, Quantum Aspects of Life, Imperial College Press, 2008, p.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, MISC, DISCIPLINE, ORGANISATION\n\nText:\nDerek Abbott, Paul C.W. Davies, and Arun K. Pati, Quantum Aspects of Life, Imperial College Press, 2008, p.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_513", "text": "The Hearthstone tournament at the 2018 Asian Games was an eSports demonstration event that was contested on 31 August 2018.", "requested_types": ["PROTEIN", "MISC", "EVENT"], "distractors": ["PROTEIN"], "gold": {"PROTEIN": [], "MISC": ["Hearthstone"], "EVENT": ["2018 Asian Games"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, EVENT\n\nText:\nThe Hearthstone tournament at the 2018 Asian Games was an eSports demonstration event that was contested on 31 August 2018.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, MISC, EVENT\n\nText:\nThe Hearthstone tournament at the 2018 Asian Games was an eSports demonstration event that was contested on 31 August 2018.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_514", "text": "In 1897 he got his Habilitation at the University of Göttingen in theoretical physics and in 1902 became extraordinary professor for theoretical physics at the University of Greifswald.", "requested_types": ["UNIVERSITY", "DISCIPLINE", "MISC"], "distractors": ["MISC"], "gold": {"UNIVERSITY": ["University of Göttingen", "University of Greifswald"], "DISCIPLINE": ["theoretical physics"], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, DISCIPLINE, MISC\n\nText:\nIn 1897 he got his Habilitation at the University of Göttingen in theoretical physics and in 1902 became extraordinary professor for theoretical physics at the University of Greifswald.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, DISCIPLINE, MISC\n\nText:\nIn 1897 he got his Habilitation at the University of Göttingen in theoretical physics and in 1902 became extraordinary professor for theoretical physics at the University of Greifswald.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_515", "text": "Mary Baker Eddy made the Boston Mother Church of Christian Science the world headquarters.", "requested_types": ["PERSON", "SCIENTIST", "LOCATION"], "distractors": ["SCIENTIST"], "gold": {"PERSON": ["Mary Baker Eddy"], "SCIENTIST": [], "LOCATION": ["Boston Mother Church of Christian Science"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, SCIENTIST, LOCATION\n\nText:\nMary Baker Eddy made the Boston Mother Church of Christian Science the world headquarters.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, SCIENTIST, LOCATION\n\nText:\nMary Baker Eddy made the Boston Mother Church of Christian Science the world headquarters.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_516", "text": "The first molecular condensates were created in November 2003 by the groups of Rudolf Grimm at the University of Innsbruck, Deborah S. Jin at the University of Colorado at Boulder and Wolfgang Ketterle at MIT.", "requested_types": ["SCIENTIST", "AWARD", "UNIVERSITY", "CHEMICAL_COMPOUND", "ENZYME"], "distractors": ["ENZYME", "CHEMICAL_COMPOUND", "AWARD"], "gold": {"SCIENTIST": ["Rudolf Grimm", "Deborah S. Jin", "Wolfgang Ketterle"], "AWARD": [], "UNIVERSITY": ["University of Innsbruck", "University of Colorado at Boulder", "MIT"], "CHEMICAL_COMPOUND": [], "ENZYME": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, UNIVERSITY, CHEMICAL_COMPOUND, ENZYME\n\nText:\nThe first molecular condensates were created in November 2003 by the groups of Rudolf Grimm at the University of Innsbruck, Deborah S. Jin at the University of Colorado at Boulder and Wolfgang Ketterle at MIT.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, AWARD, UNIVERSITY, CHEMICAL_COMPOUND, ENZYME\n\nText:\nThe first molecular condensates were created in November 2003 by the groups of Rudolf Grimm at the University of Innsbruck, Deborah S. Jin at the University of Colorado at Boulder and Wolfgang Ketterle at MIT.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_517", "text": "cAMP also, activates a protein called protein kinase A (Protein kinase A).", "requested_types": ["CHEMICAL_COMPOUND", "UNIVERSITY", "ASTRONOMICAL_OBJECT", "ENZYME", "ORGANISATION"], "distractors": ["ORGANISATION", "ASTRONOMICAL_OBJECT", "UNIVERSITY"], "gold": {"CHEMICAL_COMPOUND": ["cAMP"], "UNIVERSITY": [], "ASTRONOMICAL_OBJECT": [], "ENZYME": ["protein kinase A", "Protein kinase A"], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, UNIVERSITY, ASTRONOMICAL_OBJECT, ENZYME, ORGANISATION\n\nText:\ncAMP also, activates a protein called protein kinase A (Protein kinase A).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: CHEMICAL_COMPOUND, UNIVERSITY, ASTRONOMICAL_OBJECT, ENZYME, ORGANISATION\n\nText:\ncAMP also, activates a protein called protein kinase A (Protein kinase A).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_518", "text": "She is a fellow of the American Association for the Advancement of Science, American Chemical Society and SPIE and serves on the advisory board for ACS Nano, Advanced Functional Materials, Advanced Energy Materials, Chemical Communications, Chemistry of Materials, Materials Today, Nanoscale, and NPG Asia Materials and the board of directors for the Materials Research Society and the Polymers Materials Science and Engineering division of the American Chemical Society.", "requested_types": ["ACADEMIC_JOURNAL", "AWARD", "ORGANISATION", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"ACADEMIC_JOURNAL": ["ACS Nano", "Advanced Functional Materials", "Advanced Energy Materials", "Chemical Communications", "Chemistry of Materials", "Materials Today", "Nanoscale", "NPG Asia Materials"], "AWARD": ["fellow of the American Association for the Advancement of Science"], "ORGANISATION": ["American Chemical Society", "SPIE", "Materials Research Society", "Polymers Materials Science and Engineering division of the American Chemical Society"], "COUNTRY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, AWARD, ORGANISATION, COUNTRY\n\nText:\nShe is a fellow of the American Association for the Advancement of Science, American Chemical Society and SPIE and serves on the advisory board for ACS Nano, Advanced Functional Materials, Advanced Energy Materials, Chemical Communications, Chemistry of Materials, Materials Today, Nanoscale, and NPG Asia Materials and the board of directors for the Materials Research Society and the Polymers Materials Science and Engineering division of the American Chemical Society.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, AWARD, ORGANISATION, COUNTRY\n\nText:\nShe is a fellow of the American Association for the Advancement of Science, American Chemical Society and SPIE and serves on the advisory board for ACS Nano, Advanced Functional Materials, Advanced Energy Materials, Chemical Communications, Chemistry of Materials, Materials Today, Nanoscale, and NPG Asia Materials and the board of directors for the Materials Research Society and the Polymers Materials Science and Engineering division of the American Chemical Society.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_519", "text": "By the mid 19th century, Edward Frankland, August Kekulé, A.S. Couper, Alexander Butlerov, and Hermann Kolbe, building on the theory of radicals, developed the theory of valency, originally called combining power, in which compounds were joined owing to an attraction of positive and negative poles.", "requested_types": ["THEORY", "COUNTRY", "SCIENTIST"], "distractors": ["COUNTRY"], "gold": {"THEORY": ["theory of radicals", "theory of valency"], "COUNTRY": [], "SCIENTIST": ["Edward Frankland", "August Kekulé", "A.S. Couper", "Alexander Butlerov", "Hermann Kolbe"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, COUNTRY, SCIENTIST\n\nText:\nBy the mid 19th century, Edward Frankland, August Kekulé, A.S. Couper, Alexander Butlerov, and Hermann Kolbe, building on the theory of radicals, developed the theory of valency, originally called combining power, in which compounds were joined owing to an attraction of positive and negative poles.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: THEORY, COUNTRY, SCIENTIST\n\nText:\nBy the mid 19th century, Edward Frankland, August Kekulé, A.S. Couper, Alexander Butlerov, and Hermann Kolbe, building on the theory of radicals, developed the theory of valency, originally called combining power, in which compounds were joined owing to an attraction of positive and negative poles.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_520", "text": "This interaction stabilizes both TOC1 and PRR5 and prevents their degradation by the F-box protein ZEITLUPE (ZTL).", "requested_types": ["SCIENTIST", "COUNTRY", "PROTEIN", "MISC", "ACADEMIC_JOURNAL"], "distractors": ["COUNTRY", "SCIENTIST", "ACADEMIC_JOURNAL"], "gold": {"SCIENTIST": [], "COUNTRY": [], "PROTEIN": ["F-box protein", "ZEITLUPE", "ZTL"], "MISC": ["TOC1", "PRR5"], "ACADEMIC_JOURNAL": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, COUNTRY, PROTEIN, MISC, ACADEMIC_JOURNAL\n\nText:\nThis interaction stabilizes both TOC1 and PRR5 and prevents their degradation by the F-box protein ZEITLUPE (ZTL).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, COUNTRY, PROTEIN, MISC, ACADEMIC_JOURNAL\n\nText:\nThis interaction stabilizes both TOC1 and PRR5 and prevents their degradation by the F-box protein ZEITLUPE (ZTL).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_521", "text": "The strong lensing was predicted by Albert Einstein 's general theory of relativity and observationally discovered by Dennis Walsh, Bob Carswell, and Ray Weymann in 1979.", "requested_types": ["ACADEMIC_JOURNAL", "AWARD", "SCIENTIST", "THEORY"], "distractors": ["ACADEMIC_JOURNAL", "AWARD"], "gold": {"ACADEMIC_JOURNAL": [], "AWARD": [], "SCIENTIST": ["Albert Einstein", "Dennis Walsh", "Bob Carswell", "Ray Weymann"], "THEORY": ["general theory of relativity"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, AWARD, SCIENTIST, THEORY\n\nText:\nThe strong lensing was predicted by Albert Einstein 's general theory of relativity and observationally discovered by Dennis Walsh, Bob Carswell, and Ray Weymann in 1979.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, AWARD, SCIENTIST, THEORY\n\nText:\nThe strong lensing was predicted by Albert Einstein 's general theory of relativity and observationally discovered by Dennis Walsh, Bob Carswell, and Ray Weymann in 1979.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_522", "text": "The AAA publishes more than 20 section publications including American Anthropologist, American Ethnologist, Cultural Anthropology, Anthropology & amp; Education Quarterly and Medical Anthropology Quarterly.", "requested_types": ["PROTEIN", "SCIENTIST", "UNIVERSITY", "ACADEMIC_JOURNAL", "ORGANISATION"], "distractors": ["UNIVERSITY", "SCIENTIST", "PROTEIN"], "gold": {"PROTEIN": [], "SCIENTIST": [], "UNIVERSITY": [], "ACADEMIC_JOURNAL": ["American Anthropologist", "American Ethnologist", "Cultural Anthropology", "Anthropology", "Education Quarterly", "Medical Anthropology Quarterly"], "ORGANISATION": ["AAA"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST, UNIVERSITY, ACADEMIC_JOURNAL, ORGANISATION\n\nText:\nThe AAA publishes more than 20 section publications including American Anthropologist, American Ethnologist, Cultural Anthropology, Anthropology & amp; Education Quarterly and Medical Anthropology Quarterly.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PROTEIN, SCIENTIST, UNIVERSITY, ACADEMIC_JOURNAL, ORGANISATION\n\nText:\nThe AAA publishes more than 20 section publications including American Anthropologist, American Ethnologist, Cultural Anthropology, Anthropology & amp; Education Quarterly and Medical Anthropology Quarterly.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_523", "text": "It is one of the five Nobel Prize s established by the will of Alfred Nobel in 1895 and awarded since 1901; the others being the Nobel Prize in Chemistry, Nobel Prize in Literature, Nobel Peace Prize, and Nobel Prize in Physiology or Medicine.", "requested_types": ["SCIENTIST", "UNIVERSITY", "COUNTRY", "ACADEMIC_JOURNAL", "AWARD"], "distractors": ["COUNTRY", "ACADEMIC_JOURNAL", "UNIVERSITY"], "gold": {"SCIENTIST": ["Alfred Nobel"], "UNIVERSITY": [], "COUNTRY": [], "ACADEMIC_JOURNAL": [], "AWARD": ["Nobel Prize", "Nobel Prize in Chemistry", "Nobel Prize in Literature", "Nobel Peace Prize", "Nobel Prize in Physiology or Medicine"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, UNIVERSITY, COUNTRY, ACADEMIC_JOURNAL, AWARD\n\nText:\nIt is one of the five Nobel Prize s established by the will of Alfred Nobel in 1895 and awarded since 1901; the others being the Nobel Prize in Chemistry, Nobel Prize in Literature, Nobel Peace Prize, and Nobel Prize in Physiology or Medicine.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, UNIVERSITY, COUNTRY, ACADEMIC_JOURNAL, AWARD\n\nText:\nIt is one of the five Nobel Prize s established by the will of Alfred Nobel in 1895 and awarded since 1901; the others being the Nobel Prize in Chemistry, Nobel Prize in Literature, Nobel Peace Prize, and Nobel Prize in Physiology or Medicine.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_524", "text": "Bateman served as Executive Editor of the journal Bioinformatics from 2004 to 2012 and has also served as Editor of Nucleic Acids Research, Genome Biology and Current Protocols in Bioinformatics.", "requested_types": ["ACADEMIC_JOURNAL", "SCIENTIST", "LOCATION", "ORGANISATION"], "distractors": ["ORGANISATION", "LOCATION"], "gold": {"ACADEMIC_JOURNAL": ["Bioinformatics", "Nucleic Acids Research", "Genome Biology", "Current Protocols in Bioinformatics"], "SCIENTIST": ["Bateman"], "LOCATION": [], "ORGANISATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, LOCATION, ORGANISATION\n\nText:\nBateman served as Executive Editor of the journal Bioinformatics from 2004 to 2012 and has also served as Editor of Nucleic Acids Research, Genome Biology and Current Protocols in Bioinformatics.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ACADEMIC_JOURNAL, SCIENTIST, LOCATION, ORGANISATION\n\nText:\nBateman served as Executive Editor of the journal Bioinformatics from 2004 to 2012 and has also served as Editor of Nucleic Acids Research, Genome Biology and Current Protocols in Bioinformatics.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_525", "text": "The book on the 1902-1903 expedition contains chapters on magnetic storms on the Earth and their relationship to the Sun, the origin of the Sun itself, Halley's comet, and the rings of Saturn.", "requested_types": ["ASTRONOMICAL_OBJECT", "MISC"], "distractors": ["MISC"], "gold": {"ASTRONOMICAL_OBJECT": ["Earth", "Sun", "Halley's comet", "Saturn"], "MISC": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC\n\nText:\nThe book on the 1902-1903 expedition contains chapters on magnetic storms on the Earth and their relationship to the Sun, the origin of the Sun itself, Halley's comet, and the rings of Saturn.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, MISC\n\nText:\nThe book on the 1902-1903 expedition contains chapters on magnetic storms on the Earth and their relationship to the Sun, the origin of the Sun itself, Halley's comet, and the rings of Saturn.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_526", "text": "Phosphoinositide signaling is also altered, with elevated levels of phospholipase C, protein kinase C, and Gq alpha subunit being reported in bipolar.", "requested_types": ["MISC", "ORGANISATION", "EVENT", "ENZYME", "PERSON", "PROTEIN"], "distractors": ["ORGANISATION", "PERSON", "EVENT"], "gold": {"MISC": ["Phosphoinositide signaling"], "ORGANISATION": [], "EVENT": [], "ENZYME": ["phospholipase C", "protein kinase C"], "PERSON": [], "PROTEIN": ["Gq alpha subunit"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ORGANISATION, EVENT, ENZYME, PERSON, PROTEIN\n\nText:\nPhosphoinositide signaling is also altered, with elevated levels of phospholipase C, protein kinase C, and Gq alpha subunit being reported in bipolar.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, ORGANISATION, EVENT, ENZYME, PERSON, PROTEIN\n\nText:\nPhosphoinositide signaling is also altered, with elevated levels of phospholipase C, protein kinase C, and Gq alpha subunit being reported in bipolar.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_527", "text": "The journal has been cited most often by the following journals Journal of Clinical Pathology, Histopathology, Archives of Pathology; Laboratory Medicine, Human Pathology, and the World Journal of Gastroenterology.", "requested_types": ["COUNTRY", "ACADEMIC_JOURNAL", "ENZYME", "CHEMICAL_ELEMENT"], "distractors": ["CHEMICAL_ELEMENT", "COUNTRY", "ENZYME"], "gold": {"COUNTRY": [], "ACADEMIC_JOURNAL": ["Journal of Clinical Pathology", "Histopathology", "Archives of Pathology", "Laboratory Medicine", "Human Pathology", "World Journal of Gastroenterology"], "ENZYME": [], "CHEMICAL_ELEMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ACADEMIC_JOURNAL, ENZYME, CHEMICAL_ELEMENT\n\nText:\nThe journal has been cited most often by the following journals Journal of Clinical Pathology, Histopathology, Archives of Pathology; Laboratory Medicine, Human Pathology, and the World Journal of Gastroenterology.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, ACADEMIC_JOURNAL, ENZYME, CHEMICAL_ELEMENT\n\nText:\nThe journal has been cited most often by the following journals Journal of Clinical Pathology, Histopathology, Archives of Pathology; Laboratory Medicine, Human Pathology, and the World Journal of Gastroenterology.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_528", "text": "After important contributions of Hendrik Lorentz and Henri Poincaré, in 1905, Albert Einstein solved the problem with the introduction of special relativity, which replaced classical kinematics with a new theory of kinematics compatible with classical electromagnetism.", "requested_types": ["SCIENTIST", "THEORY", "ASTRONOMICAL_OBJECT", "MISC", "DISCIPLINE", "UNIVERSITY"], "distractors": ["MISC", "UNIVERSITY", "ASTRONOMICAL_OBJECT"], "gold": {"SCIENTIST": ["Hendrik Lorentz", "Henri Poincaré", "Albert Einstein"], "THEORY": ["special relativity"], "ASTRONOMICAL_OBJECT": [], "MISC": [], "DISCIPLINE": ["classical kinematics", "kinematics", "classical electromagnetism"], "UNIVERSITY": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, THEORY, ASTRONOMICAL_OBJECT, MISC, DISCIPLINE, UNIVERSITY\n\nText:\nAfter important contributions of Hendrik Lorentz and Henri Poincaré, in 1905, Albert Einstein solved the problem with the introduction of special relativity, which replaced classical kinematics with a new theory of kinematics compatible with classical electromagnetism.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, THEORY, ASTRONOMICAL_OBJECT, MISC, DISCIPLINE, UNIVERSITY\n\nText:\nAfter important contributions of Hendrik Lorentz and Henri Poincaré, in 1905, Albert Einstein solved the problem with the introduction of special relativity, which replaced classical kinematics with a new theory of kinematics compatible with classical electromagnetism.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_529", "text": "From 1917, he worked a few years with Niels Bohr in the University of Copenhagen and received his doctoral degree at the University College of Stockholm (now Stockholm University) in 1921.", "requested_types": ["SCIENTIST", "COUNTRY", "UNIVERSITY", "PROTEIN", "ASTRONOMICAL_OBJECT"], "distractors": ["ASTRONOMICAL_OBJECT", "COUNTRY", "PROTEIN"], "gold": {"SCIENTIST": ["Niels Bohr"], "COUNTRY": [], "UNIVERSITY": ["University of Copenhagen", "University College of Stockholm", "Stockholm University"], "PROTEIN": [], "ASTRONOMICAL_OBJECT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, COUNTRY, UNIVERSITY, PROTEIN, ASTRONOMICAL_OBJECT\n\nText:\nFrom 1917, he worked a few years with Niels Bohr in the University of Copenhagen and received his doctoral degree at the University College of Stockholm (now Stockholm University) in 1921.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SCIENTIST, COUNTRY, UNIVERSITY, PROTEIN, ASTRONOMICAL_OBJECT\n\nText:\nFrom 1917, he worked a few years with Niels Bohr in the University of Copenhagen and received his doctoral degree at the University College of Stockholm (now Stockholm University) in 1921.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_530", "text": "The film's score was composed by Harald Kloser and Thomas Wander.", "requested_types": ["PERSON", "PROTEIN"], "distractors": ["PROTEIN"], "gold": {"PERSON": ["Harald Kloser", "Thomas Wander"], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, PROTEIN\n\nText:\nThe film's score was composed by Harald Kloser and Thomas Wander.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: PERSON, PROTEIN\n\nText:\nThe film's score was composed by Harald Kloser and Thomas Wander.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_531", "text": "Type II supernovae mainly synthesize oxygen and the alpha-elements (Ne, Mg, Si, S, Ar, Ca and Ti) while Type Ia supernova e mainly produce elements of the iron peak (Ti, V, Cr, Manganese, Fe, Cobalt and Nickel) but also alpha-elements.", "requested_types": ["ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT", "THEORY", "MISC"], "distractors": ["THEORY"], "gold": {"ASTRONOMICAL_OBJECT": ["Type II supernovae", "Type Ia supernova"], "CHEMICAL_ELEMENT": ["oxygen", "Ne", "Mg", "Si", "S", "Ar", "Ca", "Ti", "V", "Cr", "Manganese", "Fe", "Cobalt", "Nickel"], "THEORY": [], "MISC": ["iron peak"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, THEORY, MISC\n\nText:\nType II supernovae mainly synthesize oxygen and the alpha-elements (Ne, Mg, Si, S, Ar, Ca and Ti) while Type Ia supernova e mainly produce elements of the iron peak (Ti, V, Cr, Manganese, Fe, Cobalt and Nickel) but also alpha-elements.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, THEORY, MISC\n\nText:\nType II supernovae mainly synthesize oxygen and the alpha-elements (Ne, Mg, Si, S, Ar, Ca and Ti) while Type Ia supernova e mainly produce elements of the iron peak (Ti, V, Cr, Manganese, Fe, Cobalt and Nickel) but also alpha-elements.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_532", "text": "More recently, some of the top awards for geophysicists are the Vetlesen Prize (intended to be the equivalent of a Nobel Prize for geology or geophysics); the William Bowie Medal (the top award of the American Geophysical Union); and the Crafoord Prize for geosciences.", "requested_types": ["DISCIPLINE", "COUNTRY", "ORGANISATION", "AWARD"], "distractors": ["COUNTRY"], "gold": {"DISCIPLINE": ["geology", "geophysics", "geosciences"], "COUNTRY": [], "ORGANISATION": ["American Geophysical Union"], "AWARD": ["Vetlesen Prize", "Nobel Prize", "William Bowie Medal", "Crafoord Prize"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, COUNTRY, ORGANISATION, AWARD\n\nText:\nMore recently, some of the top awards for geophysicists are the Vetlesen Prize (intended to be the equivalent of a Nobel Prize for geology or geophysics); the William Bowie Medal (the top award of the American Geophysical Union); and the Crafoord Prize for geosciences.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, COUNTRY, ORGANISATION, AWARD\n\nText:\nMore recently, some of the top awards for geophysicists are the Vetlesen Prize (intended to be the equivalent of a Nobel Prize for geology or geophysics); the William Bowie Medal (the top award of the American Geophysical Union); and the Crafoord Prize for geosciences.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_533", "text": "An E3 uibiquitin ligase complex composed of Cullin (CUL3), speckle-type POZ protein (SPOP), and RING-box protein 1 (RBX1) has been shown to mark the Bmi-1 with ubiquitin (a process known as ubiquitin ation).", "requested_types": ["EVENT", "PROTEIN", "MISC"], "distractors": ["EVENT"], "gold": {"EVENT": [], "PROTEIN": ["E3 uibiquitin ligase complex", "Cullin", "CUL3", "speckle-type POZ protein", "SPOP", "RING-box protein", "RBX1", "Bmi-1", "ubiquitin"], "MISC": ["ubiquitin ation"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, PROTEIN, MISC\n\nText:\nAn E3 uibiquitin ligase complex composed of Cullin (CUL3), speckle-type POZ protein (SPOP), and RING-box protein 1 (RBX1) has been shown to mark the Bmi-1 with ubiquitin (a process known as ubiquitin ation).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: EVENT, PROTEIN, MISC\n\nText:\nAn E3 uibiquitin ligase complex composed of Cullin (CUL3), speckle-type POZ protein (SPOP), and RING-box protein 1 (RBX1) has been shown to mark the Bmi-1 with ubiquitin (a process known as ubiquitin ation).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_534", "text": "He then stated that, based on his data on various comet s, the size of the Solar system can be 100 AU or even more, and that it could be other planet s there that perturb the orbit of Uranus (although the position of the eventual Neptune was not calculated until much later by Urbain Le Verrier).", "requested_types": ["ASTRONOMICAL_OBJECT", "ORGANISATION", "THEORY", "LOCATION", "SCIENTIST", "MISC"], "distractors": ["ORGANISATION", "LOCATION", "THEORY"], "gold": {"ASTRONOMICAL_OBJECT": ["Uranus", "Neptune"], "ORGANISATION": [], "THEORY": [], "LOCATION": [], "SCIENTIST": ["Urbain Le Verrier"], "MISC": ["Solar system"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ORGANISATION, THEORY, LOCATION, SCIENTIST, MISC\n\nText:\nHe then stated that, based on his data on various comet s, the size of the Solar system can be 100 AU or even more, and that it could be other planet s there that perturb the orbit of Uranus (although the position of the eventual Neptune was not calculated until much later by Urbain Le Verrier).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, ORGANISATION, THEORY, LOCATION, SCIENTIST, MISC\n\nText:\nHe then stated that, based on his data on various comet s, the size of the Solar system can be 100 AU or even more, and that it could be other planet s there that perturb the orbit of Uranus (although the position of the eventual Neptune was not calculated until much later by Urbain Le Verrier).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_535", "text": "At Institute G were Heinz Barwich, Werner Hartmann, and Justus Mühlenpfordt.", "requested_types": ["COUNTRY", "SCIENTIST", "ORGANISATION", "PROTEIN", "AWARD"], "distractors": ["AWARD", "PROTEIN", "COUNTRY"], "gold": {"COUNTRY": [], "SCIENTIST": ["Heinz Barwich", "Werner Hartmann", "Justus Mühlenpfordt"], "ORGANISATION": ["Institute G"], "PROTEIN": [], "AWARD": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, SCIENTIST, ORGANISATION, PROTEIN, AWARD\n\nText:\nAt Institute G were Heinz Barwich, Werner Hartmann, and Justus Mühlenpfordt.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: COUNTRY, SCIENTIST, ORGANISATION, PROTEIN, AWARD\n\nText:\nAt Institute G were Heinz Barwich, Werner Hartmann, and Justus Mühlenpfordt.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_536", "text": "He was a member of the Experimental Aircraft Association, the Space Pioneers, the Confederate Air Force, the Order of Daedalians, the National Rifle Association of America, the Veterans of Foreign Wars, and the Fraternal Order of Eagles.", "requested_types": ["UNIVERSITY", "COUNTRY", "ORGANISATION", "THEORY", "LOCATION"], "distractors": ["LOCATION", "UNIVERSITY", "THEORY"], "gold": {"UNIVERSITY": [], "COUNTRY": ["America"], "ORGANISATION": ["Experimental Aircraft Association", "Space Pioneers", "Confederate Air Force", "Order of Daedalians", "National Rifle Association", "Veterans of Foreign Wars", "Fraternal Order of Eagles"], "THEORY": [], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, COUNTRY, ORGANISATION, THEORY, LOCATION\n\nText:\nHe was a member of the Experimental Aircraft Association, the Space Pioneers, the Confederate Air Force, the Order of Daedalians, the National Rifle Association of America, the Veterans of Foreign Wars, and the Fraternal Order of Eagles.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, COUNTRY, ORGANISATION, THEORY, LOCATION\n\nText:\nHe was a member of the Experimental Aircraft Association, the Space Pioneers, the Confederate Air Force, the Order of Daedalians, the National Rifle Association of America, the Veterans of Foreign Wars, and the Fraternal Order of Eagles.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_537", "text": "Overall Bamberga is the tenth-brightest main-belt asteroid after, in order, 4 Vesta, 2 Pallas, Ceres, 7 Iris, 6 Hebe, 3 Juno, 18 Melpomene, 15 Eunomia and 8 Flora.", "requested_types": ["ASTRONOMICAL_OBJECT", "LOCATION", "DISCIPLINE", "CHEMICAL_COMPOUND"], "distractors": ["DISCIPLINE", "CHEMICAL_COMPOUND", "LOCATION"], "gold": {"ASTRONOMICAL_OBJECT": ["Bamberga", "4 Vesta", "2 Pallas", "Ceres", "7 Iris", "6 Hebe", "3 Juno", "18 Melpomene", "15 Eunomia", "8 Flora"], "LOCATION": [], "DISCIPLINE": [], "CHEMICAL_COMPOUND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, LOCATION, DISCIPLINE, CHEMICAL_COMPOUND\n\nText:\nOverall Bamberga is the tenth-brightest main-belt asteroid after, in order, 4 Vesta, 2 Pallas, Ceres, 7 Iris, 6 Hebe, 3 Juno, 18 Melpomene, 15 Eunomia and 8 Flora.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, LOCATION, DISCIPLINE, CHEMICAL_COMPOUND\n\nText:\nOverall Bamberga is the tenth-brightest main-belt asteroid after, in order, 4 Vesta, 2 Pallas, Ceres, 7 Iris, 6 Hebe, 3 Juno, 18 Melpomene, 15 Eunomia and 8 Flora.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_538", "text": "The staff also produce articles on human and animal genetics topics such as gene splicing, CRISPR, government regulation, bioethics, use of stem cells, transhumanism, nanotechnology and synthetic biology.", "requested_types": ["MISC", "DISCIPLINE", "SCIENTIST", "THEORY", "LOCATION"], "distractors": ["LOCATION", "SCIENTIST"], "gold": {"MISC": ["gene splicing", "CRISPR", "government regulation", "stem cells"], "DISCIPLINE": ["genetics", "bioethics", "nanotechnology", "synthetic biology"], "SCIENTIST": [], "THEORY": ["transhumanism"], "LOCATION": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, DISCIPLINE, SCIENTIST, THEORY, LOCATION\n\nText:\nThe staff also produce articles on human and animal genetics topics such as gene splicing, CRISPR, government regulation, bioethics, use of stem cells, transhumanism, nanotechnology and synthetic biology.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MISC, DISCIPLINE, SCIENTIST, THEORY, LOCATION\n\nText:\nThe staff also produce articles on human and animal genetics topics such as gene splicing, CRISPR, government regulation, bioethics, use of stem cells, transhumanism, nanotechnology and synthetic biology.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_539", "text": "He has won awards from the American Psychological Association, the National Academy of Sciences, the Royal Institution, the Cognitive Neuroscience Society and the American Humanist Association.", "requested_types": ["UNIVERSITY", "COUNTRY", "ORGANISATION"], "distractors": ["UNIVERSITY", "COUNTRY"], "gold": {"UNIVERSITY": [], "COUNTRY": [], "ORGANISATION": ["American Psychological Association", "National Academy of Sciences", "Royal Institution", "Cognitive Neuroscience Society", "American Humanist Association"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, COUNTRY, ORGANISATION\n\nText:\nHe has won awards from the American Psychological Association, the National Academy of Sciences, the Royal Institution, the Cognitive Neuroscience Society and the American Humanist Association.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: UNIVERSITY, COUNTRY, ORGANISATION\n\nText:\nHe has won awards from the American Psychological Association, the National Academy of Sciences, the Royal Institution, the Cognitive Neuroscience Society and the American Humanist Association.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_540", "text": "Current research in the Liang Tong's lab focuses on enzymes involved in fatty acid metabolism, including Acetyl-CoA carboxylase, carnitine acyltransferase, AMP-activated protein kinase, and others.", "requested_types": ["AWARD", "SCIENTIST", "ENZYME", "CHEMICAL_COMPOUND", "ACADEMIC_JOURNAL", "PROTEIN"], "distractors": ["PROTEIN", "AWARD", "ACADEMIC_JOURNAL"], "gold": {"AWARD": [], "SCIENTIST": ["Liang Tong"], "ENZYME": ["Acetyl-CoA carboxylase", "carnitine acyltransferase", "AMP-activated protein kinase"], "CHEMICAL_COMPOUND": ["fatty acid"], "ACADEMIC_JOURNAL": [], "PROTEIN": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, SCIENTIST, ENZYME, CHEMICAL_COMPOUND, ACADEMIC_JOURNAL, PROTEIN\n\nText:\nCurrent research in the Liang Tong's lab focuses on enzymes involved in fatty acid metabolism, including Acetyl-CoA carboxylase, carnitine acyltransferase, AMP-activated protein kinase, and others.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: AWARD, SCIENTIST, ENZYME, CHEMICAL_COMPOUND, ACADEMIC_JOURNAL, PROTEIN\n\nText:\nCurrent research in the Liang Tong's lab focuses on enzymes involved in fatty acid metabolism, including Acetyl-CoA carboxylase, carnitine acyltransferase, AMP-activated protein kinase, and others.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_541", "text": "The main stadium of the city is the Kaftanzoglio Stadium (also home ground of Iraklis FC), while other main stadiums of the city include the football Kleanthis Vikelidis Stadium and Toumba Stadium home grounds of Aris F.C. and PAOK F.C., respectively, all of whom are founding members of the Greek league.", "requested_types": ["ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT", "MISC", "ORGANISATION", "LOCATION"], "distractors": ["ASTRONOMICAL_OBJECT", "CHEMICAL_ELEMENT"], "gold": {"ASTRONOMICAL_OBJECT": [], "CHEMICAL_ELEMENT": [], "MISC": ["Greek league"], "ORGANISATION": ["Iraklis FC", "Aris F.C.", "PAOK F.C."], "LOCATION": ["Kaftanzoglio Stadium", "Kleanthis Vikelidis Stadium", "Toumba Stadium"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, MISC, ORGANISATION, LOCATION\n\nText:\nThe main stadium of the city is the Kaftanzoglio Stadium (also home ground of Iraklis FC), while other main stadiums of the city include the football Kleanthis Vikelidis Stadium and Toumba Stadium home grounds of Aris F.C. and PAOK F.C., respectively, all of whom are founding members of the Greek league.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ASTRONOMICAL_OBJECT, CHEMICAL_ELEMENT, MISC, ORGANISATION, LOCATION\n\nText:\nThe main stadium of the city is the Kaftanzoglio Stadium (also home ground of Iraklis FC), while other main stadiums of the city include the football Kleanthis Vikelidis Stadium and Toumba Stadium home grounds of Aris F.C. and PAOK F.C., respectively, all of whom are founding members of the Greek league.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_science_542", "text": "Mattauch and Fritz Strassmann actively supported the proposed appointment of Lise Meitner as head of the physics department of the University of Mainz.", "requested_types": ["DISCIPLINE", "SCIENTIST", "CHEMICAL_ELEMENT", "ORGANISATION"], "distractors": ["CHEMICAL_ELEMENT", "DISCIPLINE"], "gold": {"DISCIPLINE": [], "SCIENTIST": ["Mattauch", "Fritz Strassmann", "Lise Meitner"], "CHEMICAL_ELEMENT": [], "ORGANISATION": ["physics department of the University of Mainz"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, SCIENTIST, CHEMICAL_ELEMENT, ORGANISATION\n\nText:\nMattauch and Fritz Strassmann actively supported the proposed appointment of Lise Meitner as head of the physics department of the University of Mainz.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: DISCIPLINE, SCIENTIST, CHEMICAL_ELEMENT, ORGANISATION\n\nText:\nMattauch and Fritz Strassmann actively supported the proposed appointment of Lise Meitner as head of the physics department of the University of Mainz.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}]