Dataset Viewer
Auto-converted to Parquet Duplicate
system_prompt
stringclasses
1 value
user_prompt
stringlengths
498
793
response
stringlengths
78
346
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Amityville is the westernmost station on the Babylon Branch of the Long Island Rail Road in Suffolk County , New York . </input> Subject entity: Babylon Branch Object entity: Long Island Rail Road Your task is to identify the relation that holds between the subject entity and the object ...
{"events": [["Babylon Branch", "owned by", [["Long Island Rail Road", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In 2017 Amazon purchased Whole Foods and began selling 365 ( whole foods store brand products ) through Amazon fresh . </input> Subject entity: Whole Foods Object entity: Amazon Your task is to identify the relation that holds between the subject entity and the object entity. For each r...
{"events": [["Whole Foods", "owned by", [["Amazon", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> A member of the senior Spanish national team , Royal Spanish Football Federation she scored against England and Northern Ireland at the UEFA Women 's Euro 2009 qualifying . </input> Subject entity: Spanish national team Object entity: Royal Spanish Football Federation Your task is to ide...
{"events": [["Spanish national team", "owned by", [["Royal Spanish Football Federation", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Cannes hosts the Cannes Mandelieu Space Center , headquarters of Thales Alenia Space , the first European satellite manufacturer . </input> Subject entity: Cannes Mandelieu Space Center Object entity: Thales Alenia Space Your task is to identify the relation that holds between the subjec...
{"events": [["Cannes Mandelieu Space Center", "owned by", [["Thales Alenia Space", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In August 2014 , SmartThings announced that they had reached an agreement to be acquired by Samsung and would operate as an independent company within Samsung 's Open Innovation Center . </input> Subject entity: SmartThings Object entity: Samsung Your task is to identify the relation tha...
{"events": [["SmartThings", "owned by", [["Samsung", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The Keihan Electric Railway was at that time constructing the Ōtō Line to the Eiden terminal at Demachiyanagi . </input> Subject entity: Ōtō Line Object entity: Keihan Electric Railway Your task is to identify the relation that holds between the subject entity and the object entity. For...
{"events": [["\u014ct\u014d Line", "owned by", [["Keihan Electric Railway", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> " La Meuse " has its headquarters in Liège and is owned by the Rossel group which also owns " Le Soir " and " La Lanterne " , among the others . </input> Subject entity: Le Soir Object entity: Rossel group Your task is to identify the relation that holds between the subject entity and th...
{"events": [["Le Soir", "owned by", [["Rossel group", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In 2005 , Britvic commissioned Furr to create a painting marking 70 years of Robinsons ( drink ) the Wimbledon Championships on Centre Court . </input> Subject entity: Robinsons ( drink ) Object entity: Britvic Your task is to identify the relation that holds between the subject entity a...
{"events": [["Robinsons ( drink )", "owned by", [["Britvic", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The school is located in the city of Mannheim , Baden - Württemberg in Germany at Mannheim Palace , one of the largest baroque castles in Europe . </input> Subject entity: Mannheim Palace Object entity: Baden - Württemberg Your task is to identify the relation that holds between the subj...
{"events": [["Mannheim Palace", "owned by", [["Baden - W\u00fcrttemberg", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> One development occurred in mid 2013 , when billionaire Jeff Bezos , founder of Amazon.com , paid US$ 250 million for " The Washington Post " and several smaller newspapers . </input> Subject entity: The Washington Post Object entity: Jeff Bezos Your task is to identify the relation that...
{"events": [["The Washington Post", "owned by", [["Jeff Bezos", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In February 2018 , TUI Cruises ordered a new ship , Mein Schiff 7 , that will join the company 's fleet in 2023 . </input> Subject entity: Mein Schiff Object entity: TUI Cruises Your task is to identify the relation that holds between the subject entity and the object entity. For each r...
{"events": [["Mein Schiff", "owned by", [["TUI Cruises", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The station forms part of Transport for New South Wales ' Sydney Metro City & Southwest scheme . </input> Subject entity: Sydney Metro City & Southwest Object entity: Transport for New South Wales Your task is to identify the relation that holds between the subject entity and the object ...
{"events": [["Sydney Metro City & Southwest", "owned by", [["Transport for New South Wales", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In October of the same year , the Delhi High Court ordered it to stop publishing articles from " Mint " after " Mint " owner HT Media filed a copyright infringement suit . </input> Subject entity: Mint Object entity: HT Media Your task is to identify the relation that holds between the s...
{"events": [["Mint", "owned by", [["HT Media", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The first ever watch to be sent into space was a Russian " Pobeda " watch from the Petrodvorets Watch Factory . </input> Subject entity: Pobeda Object entity: Petrodvorets Watch Factory Your task is to identify the relation that holds between the subject entity and the object entity. Fo...
{"events": [["Pobeda", "owned by", [["Petrodvorets Watch Factory", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Vivendi went on to acquire stakes in or merge with Maroc Telecom , Havas , Cendant Software , Grupo Anaya , and NetHold , a large Continental European pay - TV operator . </input> Subject entity: Havas Object entity: Vivendi Your task is to identify the relation that holds between the su...
{"events": [["Havas", "owned by", [["Vivendi", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In 2009 , Groupe PSA created a new brand - DS Automobiles , intended as high quality , high specification variations on existing models , with differing mechanics and bodywork . </input> Subject entity: DS Automobiles Object entity: Groupe PSA Your task is to identify the relation that h...
{"events": [["DS Automobiles", "owned by", [["Groupe PSA", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> By 1967 , with the manor house gone and the garden derelict , the National Trust — who had never been responsible for a garden restoration before — accepted the care of Westbury Court Garden . </input> Subject entity: Westbury Court Garden Object entity: National Trust Your task is to id...
{"events": [["Westbury Court Garden", "owned by", [["National Trust", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Savant Lake railway station is on the line and is served by Via Rail transcontinental " Canadian " trains . </input> Subject entity: Savant Lake railway station Object entity: Via Rail Your task is to identify the relation that holds between the subject entity and the object entity. For...
{"events": [["Savant Lake railway station", "owned by", [["Via Rail", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Rostokino ( ) is a station on the Moscow Central Circle of the Moscow Metro that opened in September 2016 . </input> Subject entity: Moscow Central Circle Object entity: Moscow Metro Your task is to identify the relation that holds between the subject entity and the object entity. For e...
{"events": [["Moscow Central Circle", "owned by", [["Moscow Metro", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Currently , CBS Corporation owns the rights to the film via its Simon & Schuster subsidiary , who owns the rights to the Raggedy Ann and Andy characters . </input> Subject entity: Simon & Schuster Object entity: CBS Corporation Your task is to identify the relation that holds between the...
{"events": [["Simon & Schuster", "owned by", [["CBS Corporation", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> It is the junction point of the Korean State Railway 's P'yŏngra and Ŭnsan lines . </input> Subject entity: Ŭnsan Object entity: Korean State Railway Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the rela...
{"events": [["\u016cnsan", "owned by", [["Korean State Railway", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> " Fiction Factory " ( in Spanish : Factoría de Ficción ) styled as FDF , is a private Spanish television network owned by Mediaset España Comunicación . </input> Subject entity: Factoría de Ficción Object entity: Mediaset España Comunicación Your task is to identify the relation that hol...
{"events": [["Factor\u00eda de Ficci\u00f3n", "owned by", [["Mediaset Espa\u00f1a Comunicaci\u00f3n", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The WSU – Idaho game in 1970 was dubbed the Displaced Bowl , which was held in Joe Albi Stadium in Spokane on September 19 . </input> Subject entity: Joe Albi Stadium Object entity: Spokane Your task is to identify the relation that holds between the subject entity and the object entity....
{"events": [["Joe Albi Stadium", "owned by", [["Spokane", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The mall is located less than a mile away from the Northshore Mall , in Peabody , although both malls are primarily owned and managed by Simon Property Group . </input> Subject entity: Northshore Mall Object entity: Simon Property Group Your task is to identify the relation that holds be...
{"events": [["Northshore Mall", "owned by", [["Simon Property Group", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> On 1 May 2006 , Siddhartha Lal took over as the Chief executive officer and Managing Director of Eicher Motors . </input> Subject entity: Eicher Motors Object entity: Siddhartha Lal Your task is to identify the relation that holds between the subject entity and the object entity. For ea...
{"events": [["Eicher Motors", "owned by", [["Siddhartha Lal", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> No other office skyscraper was built in Philadelphia until Brandywine Realty Trust ( NYSE : BDN ) built the Cira Centre in 2005 . </input> Subject entity: Cira Centre Object entity: Brandywine Realty Trust Your task is to identify the relation that holds between the subject entity and th...
{"events": [["Cira Centre", "owned by", [["Brandywine Realty Trust", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Portugal Telecom is no longer a publicly traded company , since 100 % of its capital is owned by Altice Group . </input> Subject entity: Portugal Telecom Object entity: Altice Group Your task is to identify the relation that holds between the subject entity and the object entity. For ea...
{"events": [["Portugal Telecom", "owned by", [["Altice Group", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Thomson Francis Mason was originally interred at the Colross graveyard in 1838 , as were two of his daughters . </input> Subject entity: Colross Object entity: Thomson Francis Mason Your task is to identify the relation that holds between the subject entity and the object entity. For ea...
{"events": [["Colross", "owned by", [["Thomson Francis Mason", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Gorilla is a British advertising campaign launched by Cadbury Schweppes in 2007 , to promote Cadbury Dairy Milk - brand chocolate . </input> Subject entity: Cadbury Dairy Milk Object entity: Cadbury Schweppes Your task is to identify the relation that holds between the subject entity and...
{"events": [["Cadbury Dairy Milk", "owned by", [["Cadbury Schweppes", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The Prasar Bharati board approved the proposal to start a 24-hour news channel in place of DD Metro , which was closing . </input> Subject entity: DD Metro Object entity: Prasar Bharati Your task is to identify the relation that holds between the subject entity and the object entity. Fo...
{"events": [["DD Metro", "owned by", [["Prasar Bharati", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The station was included in a nation - wide program of upgrades to the main Italian stations , by Grandi Stazioni , a subsidiary of Ferrovie dello Stato . </input> Subject entity: Grandi Stazioni Object entity: Ferrovie dello Stato Your task is to identify the relation that holds between...
{"events": [["Grandi Stazioni", "owned by", [["Ferrovie dello Stato", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> London River Services operate boats from nearby London Eye Pier ( also known as the Waterloo Millennium Pier ) and Festival Pier , and run to the City and Greenwich . </input> Subject entity: Festival Pier Object entity: London River Services Your task is to identify the relation that ho...
{"events": [["Festival Pier", "owned by", [["London River Services", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> However , in 1992 , then - owner Aamulehti sold Pohjalainen to Ilkka . </input> Subject entity: Pohjalainen Object entity: Ilkka Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the relation type - Descripti...
{"events": [["Pohjalainen", "owned by", [["Ilkka", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Originally built entirely as a narrow gauge line , frequent accidents on the line led the Korean State Railway to convert the Hamhŭng — Sinhŭng to standard gauge for greater safety and increased transportation capacity . </input> Subject entity: Sinhŭng Object entity: Korean State Railway...
{"events": [["Sinh\u016dng", "owned by", [["Korean State Railway", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The plan is foiled when Samwise Gamgee greatly injures Shelob with Frodo 's Elvish dagger , Sting , and the Phial of Galadriel . </input> Subject entity: Sting Object entity: Frodo Your task is to identify the relation that holds between the subject entity and the object entity. For eac...
{"events": [["Sting", "owned by", [["Frodo", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The Milton Hershey School Trust also has 100 % control of Hershey Entertainment and Resorts Company , which owns the Hotel Hershey and Hersheypark , among other properties . </input> Subject entity: Hershey Entertainment and Resorts Company Object entity: Milton Hershey School Trust Your...
{"events": [["Hershey Entertainment and Resorts Company", "owned by", [["Milton Hershey School Trust", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> He has also been the CEO for McCann - Erickson and MTG which has the channels TV3 , TV6 ad ZTV . </input> Subject entity: TV3 Object entity: MTG Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the relation ...
{"events": [["TV3", "owned by", [["MTG", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Until March 2014 , it was 21 % owned by Kingfisher plc , a UK company , who also own the B&Q ; and Castorama DIY chains . </input> Subject entity: Castorama Object entity: Kingfisher plc Your task is to identify the relation that holds between the subject entity and the object entity. F...
{"events": [["Castorama", "owned by", [["Kingfisher plc", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> He bought Wakehurst Place , now owned by the National Trust in 1938 . </input> Subject entity: Wakehurst Place Object entity: National Trust Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the relation type...
{"events": [["Wakehurst Place", "owned by", [["National Trust", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Powers of the Vancouver council include : City council sits at City Hall at 453 West 12th Avenue . </input> Subject entity: City Hall Object entity: Vancouver Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Iden...
{"events": [["City Hall", "owned by", [["Vancouver", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Viacom International Media Networks Europe launched its first channel MTV Europe on 1 August 1987 from Amsterdam , with its main headquarters in London . </input> Subject entity: MTV Europe Object entity: Viacom International Media Networks Europe Your task is to identify the relation th...
{"events": [["MTV Europe", "owned by", [["Viacom International Media Networks Europe", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Originally operating as part of the D Line , the station was opened on December 19 , 1995 , and is operated by the Regional Transportation District . </input> Subject entity: D Line Object entity: Regional Transportation District Your task is to identify the relation that holds between t...
{"events": [["D Line", "owned by", [["Regional Transportation District", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> For the commercial Eric Clapton brought his famous black Fender Stratocaster , " Blackie " , out of retirement at the request of Honda Cars in Japan . </input> Subject entity: Blackie Object entity: Eric Clapton Your task is to identify the relation that holds between the subject entity ...
{"events": [["Blackie", "owned by", [["Eric Clapton", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> On December 21 , 2017 it was announced that the Buffalo Beauts had been acquired by Pegula Sports and Entertainment , owners of the Buffalo Sabres , Rochester Americans and the Beauts arena HarborCenter . </input> Subject entity: Buffalo Beauts Object entity: Pegula Sports and Entertainme...
{"events": [["Buffalo Beauts", "owned by", [["Pegula Sports and Entertainment", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Othello station is part of Sound Transit 's Central Link line , which runs from Seattle – Tacoma International Airport through the Rainier Valley and Downtown Seattle to the University of Washington . </input> Subject entity: Central Link Object entity: Sound Transit Your task is to iden...
{"events": [["Central Link", "owned by", [["Sound Transit", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The nearest passenger rail station to New Castle is Wilmington station in Wilmington , which is served by Amtrak and SEPTA Regional Rail 's Wilmington / Newark Line . </input> Subject entity: Wilmington station Object entity: Amtrak Your task is to identify the relation that holds betwee...
{"events": [["Wilmington station", "owned by", [["Amtrak", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> ' Dream Mall ' is a joint development project between Uni - President Corp. and Tainan Spinning Ltd. </input> Subject entity: Dream Mall Object entity: Uni - President Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: -...
{"events": [["Dream Mall", "owned by", [["Uni - President", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Rendlesham Forest is owned by the Forestry Commission and consists of about of coniferous plantations , interspersed with broadleaved belts , heathland and wetland areas . </input> Subject entity: Rendlesham Forest Object entity: Forestry Commission Your task is to identify the relation ...
{"events": [["Rendlesham Forest", "owned by", [["Forestry Commission", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In December 2010 , Adevărul Holding also launched a sister version of its title asset , published in neighboring Moldova as " Adevărul Moldova " . </input> Subject entity: Adevărul Moldova Object entity: Adevărul Holding Your task is to identify the relation that holds between the subjec...
{"events": [["Adev\u0103rul Moldova", "owned by", [["Adev\u0103rul Holding", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The line forms part of what JR East refers to as the around Tokyo , consisting of the Keiyo Line , Musashino Line , Nambu Line , and Yokohama Line . </input> Subject entity: Nambu Line Object entity: JR East Your task is to identify the relation that holds between the subject entity and ...
{"events": [["Nambu Line", "owned by", [["JR East", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> April 11 , 2011.Building the World ’s First Mobile Rewards Network : Brian Wong , 20-Year - Old Founder of Kiip . </input> Subject entity: Kiip Object entity: Brian Wong Your task is to identify the relation that holds between the subject entity and the object entity. For each relation:...
{"events": [["Kiip", "owned by", [["Brian Wong", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> News Corporation and Telstra created the rival Foxtel consortium . </input> Subject entity: Foxtel Object entity: News Corporation Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the relation type - Descrip...
{"events": [["Foxtel", "owned by", [["News Corporation", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> It was bought by NECCO in 1999.Pittsburgh ’s Very Own , Vintage Clark Bar and Zagnut Bar at CandyFavorites.com candy blog </input> Subject entity: Clark Bar Object entity: NECCO Your task is to identify the relation that holds between the subject entity and the object entity. For each r...
{"events": [["Clark Bar", "owned by", [["NECCO", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Formed in 2010 by Haim Saban , Saban Capital Group owns Saban Brands , part of Univision Communications , and part of Celestial Tiger Entertainment . </input> Subject entity: Celestial Tiger Entertainment Object entity: Saban Capital Group Your task is to identify the relation that holds...
{"events": [["Celestial Tiger Entertainment", "owned by", [["Saban Capital Group", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The Cowley plant is now where BMW 's Mini is assembled , known At Plant Oxford . </input> Subject entity: Mini Object entity: BMW Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the relation type - Descript...
{"events": [["Mini", "owned by", [["BMW", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Along and around the South Bank , a tourist area has developed around the former Greater London Council headquarters of County Hall and the Southbank Centre and National Theatre . </input> Subject entity: County Hall Object entity: Greater London Council Your task is to identify the rela...
{"events": [["County Hall", "owned by", [["Greater London Council", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The Spear symbolizes the lance of Longinus that was used at the time of Christ 's death on the Cross to pierce his side , from which flowed blood and water ( ) . </input> Subject entity: lance Object entity: Longinus Your task is to identify the relation that holds between the subject en...
{"events": [["lance", "owned by", [["Longinus", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The Panamanian Football Federation ( ) , known as FEPAFUT , is the official governing body of football in Panama and is in charge of the Panama national football team . </input> Subject entity: Panama national football team Object entity: FEPAFUT Your task is to identify the relation tha...
{"events": [["Panama national football team", "owned by", [["FEPAFUT", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The Røa Line is owned by Kollektivtransportproduksjon , and operated by Oslo T - banedrift on contract with the public transport agency Ruter . </input> Subject entity: Røa Object entity: Kollektivtransportproduksjon Your task is to identify the relation that holds between the subject en...
{"events": [["R\u00f8a", "owned by", [["Kollektivtransportproduksjon", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The Nine Network 's owner , PBL Media , purchased affiliate station NBN Television from its owners SP Telemedia , in May 2007 . </input> Subject entity: Nine Network Object entity: PBL Media Your task is to identify the relation that holds between the subject entity and the object entity...
{"events": [["Nine Network", "owned by", [["PBL Media", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Tata Prima is a range of heavy trucks produced by Tata Daewoo , a wholly owned subsidiary of Tata Motors of India . </input> Subject entity: Tata Daewoo Object entity: Tata Motors Your task is to identify the relation that holds between the subject entity and the object entity. For each...
{"events": [["Tata Daewoo", "owned by", [["Tata Motors", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> On 28 March 2010 , Geely signed a deal worth US$ 1.8 billion to buy Volvo Cars from American automobile manufacturer Ford Motor Company . </input> Subject entity: Volvo Cars Object entity: Geely Your task is to identify the relation that holds between the subject entity and the object en...
{"events": [["Volvo Cars", "owned by", [["Geely", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> His memorial service was held on December 30 , 2004 , at the Baughman Center that he and Hazel had given to the University of Florida . </input> Subject entity: Baughman Center Object entity: University of Florida Your task is to identify the relation that holds between the subject entit...
{"events": [["Baughman Center", "owned by", [["University of Florida", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Devassa is a brand of beer based in Brazil made by Brasil Kirin under license from the U.S. company Playboy . </input> Subject entity: Devassa Object entity: Brasil Kirin Your task is to identify the relation that holds between the subject entity and the object entity. For each relation...
{"events": [["Devassa", "owned by", [["Brasil Kirin", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The following is a list of species recognised by Tropicos maintained by the Missouri Botanical Garden : </input> Subject entity: Tropicos Object entity: Missouri Botanical Garden Your task is to identify the relation that holds between the subject entity and the object entity. For each ...
{"events": [["Tropicos", "owned by", [["Missouri Botanical Garden", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> On 14 October 2011 , KSÍ announced they had appointed Heimir as assistant manager of the Iceland national football team as they announced they had appointed Lars Lagerbäck as manager . </input> Subject entity: Iceland national football team Object entity: KSÍ Your task is to identify the...
{"events": [["Iceland national football team", "owned by", [["KS\u00cd", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Originally known as " Sendai distillery " , it was the second to be established by Nikka Whisky , after the company ’s Yoichi distillery in Hokkaido . </input> Subject entity: Yoichi distillery Object entity: Nikka Whisky Your task is to identify the relation that holds between the subje...
{"events": [["Yoichi distillery", "owned by", [["Nikka Whisky", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Canwest ultimately went bankrupt in 2010 , selling the bulk of its broadcast properties including Global BC to Corus sibling company Shaw Communications , which in turn transferred them to Corus in 2016 . </input> Subject entity: Global BC Object entity: Shaw Communications Your task is ...
{"events": [["Global BC", "owned by", [["Shaw Communications", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Sion play their home matches at the multi - purpose Stade Tourbillon , which has a capacity of 14,500 . </input> Subject entity: Stade Tourbillon Object entity: Sion Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - I...
{"events": [["Stade Tourbillon", "owned by", [["Sion", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Khan joined Mikhail Fridman in establishing LetterOne , based in Luxembourg , for the purpose of investing proceeds of the proceeds from the TNK - BP sale in international projects . </input> Subject entity: LetterOne Object entity: Mikhail Fridman Your task is to identify the relation t...
{"events": [["LetterOne", "owned by", [["Mikhail Fridman", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In April 2016 KBL " epb " has signed a preliminary agreement to acquire Insinger de Beaufort from the parent BNP Paribas Wealth Management . </input> Subject entity: Insinger de Beaufort Object entity: BNP Paribas Wealth Management Your task is to identify the relation that holds between...
{"events": [["Insinger de Beaufort", "owned by", [["BNP Paribas Wealth Management", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In 2015 Trinity Mirror bought Local World 's newspapers and online services . </input> Subject entity: Local World Object entity: Trinity Mirror Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the relation ...
{"events": [["Local World", "owned by", [["Trinity Mirror", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In August 2015 , GrafTech was purchased by Brookfield Asset Management . </input> Subject entity: GrafTech Object entity: Brookfield Asset Management Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the rela...
{"events": [["GrafTech", "owned by", [["Brookfield Asset Management", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> From just after Finchley Road these four tracks run parallel with the Network Rail Chiltern Main Line from Marylebone . </input> Subject entity: Chiltern Main Line Object entity: Network Rail Your task is to identify the relation that holds between the subject entity and the object entit...
{"events": [["Chiltern Main Line", "owned by", [["Network Rail", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The " Keighley News " is owned by Newsquest , the second largest publisher of regional newspapers in the United Kingdom . </input> Subject entity: Keighley News Object entity: Newsquest Your task is to identify the relation that holds between the subject entity and the object entity. Fo...
{"events": [["Keighley News", "owned by", [["Newsquest", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> From 2016 , " Morgenbladet " 's headquarters are co - located with NHST Media Group and Dagens Næringsliv at Akerselva Atrium . </input> Subject entity: Morgenbladet Object entity: NHST Media Group Your task is to identify the relation that holds between the subject entity and the object...
{"events": [["Morgenbladet", "owned by", [["NHST Media Group", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> On June 2 , 2013 , AOL sold Noisecreep , The Boot and The Boombox to Townsquare Media . AOL ’s Orphaned Music Sites Find a Home at Townsquare Media </input> Subject entity: Noisecreep Object entity: Townsquare Media Your task is to identify the relation that holds between the subject ent...
{"events": [["Noisecreep", "owned by", [["Townsquare Media", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> On March 6 , 2014 Spotify announced that they had acquired The Echo Nest . </input> Subject entity: The Echo Nest Object entity: Spotify Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the relation type - D...
{"events": [["The Echo Nest", "owned by", [["Spotify", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Dallas Mavericks owner Mark Cuban had earlier announced that Harper 's jersey was to be retired during the 2013–2014 season , but later Cuban said that it would not be retired that season . </input> Subject entity: Dallas Mavericks Object entity: Mark Cuban Your task is to identify the r...
{"events": [["Dallas Mavericks", "owned by", [["Mark Cuban", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> He learned that the Italian Line 's SS " Rex " was bound for New York and would pass the mark on 11 May. </input> Subject entity: SS " Rex " Object entity: Italian Line Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: ...
{"events": [["SS \" Rex \"", "owned by", [["Italian Line", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The construction of Edo Castle by Ōta Dōkan , a vassal of Uesugi Mochitomo , began in 1457 during the Muromachi period in what is now the East Garden of the Imperial Palace . </input> Subject entity: Edo Castle Object entity: Ōta Dōkan Your task is to identify the relation that holds bet...
{"events": [["Edo Castle", "owned by", [["\u014cta D\u014dkan", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Super VHS - C or S - VHS Compact was developed by JVC in 1987 . </input> Subject entity: Super VHS Object entity: JVC Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the relation type - Description: explain...
{"events": [["Super VHS", "owned by", [["JVC", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> On Saturday 27 October 1951 , 33,230 spectators saw Leigh lose 14–6 to Wigan in the final of the Lancashire County Cup at Station Road , Swinton . </input> Subject entity: Station Road Object entity: Swinton Your task is to identify the relation that holds between the subject entity and ...
{"events": [["Station Road", "owned by", [["Swinton", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> College Hall at the University of Pennsylvania , for example , is constructed out of serpentine . </input> Subject entity: College Hall Object entity: University of Pennsylvania Your task is to identify the relation that holds between the subject entity and the object entity. For each r...
{"events": [["College Hall", "owned by", [["University of Pennsylvania", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> For a time it was also California 's tallest building , replacing both the Russ Building in San Francisco and the Los Angeles City Hall in Los Angeles , California . </input> Subject entity: Los Angeles City Hall Object entity: Los Angeles , California Your task is to identify the relati...
{"events": [["Los Angeles City Hall", "owned by", [["Los Angeles , California", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The station is owned by the Nation Broadcasting Corporation , a subsidiary of PLDT Beneficial Trust Fund 's broadcasting division MediaQuest Holdings . </input> Subject entity: Nation Broadcasting Corporation Object entity: MediaQuest Holdings Your task is to identify the relation that h...
{"events": [["Nation Broadcasting Corporation", "owned by", [["MediaQuest Holdings", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The adjacent Ogden Intermodal Transit Center currently serves the Utah Transit Authority 's ( UTA ) " FrontRunner " commuter rail line . </input> Subject entity: Ogden Intermodal Transit Center Object entity: Utah Transit Authority Your task is to identify the relation that holds between...
{"events": [["Ogden Intermodal Transit Center", "owned by", [["Utah Transit Authority", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Its sister newspapers include " ABC " , " El Correo Español " , " El Diario Vasco " and " La Verdad " , all of which are part of the Vocento Group . </input> Subject entity: ABC Object entity: Vocento Group Your task is to identify the relation that holds between the subject entity and t...
{"events": [["ABC", "owned by", [["Vocento Group", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> He co - founded Banco Safra with his brothers Edmond Safra and Joseph Safra . </input> Subject entity: Banco Safra Object entity: Joseph Safra Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the relation ty...
{"events": [["Banco Safra", "owned by", [["Joseph Safra", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The magazine was published by ABP Group , whose most prominent publications are " The Telegraph " , Anandabazar Patrika , Sananda , Anandamela and others . </input> Subject entity: Anandabazar Patrika Object entity: ABP Group Your task is to identify the relation that holds between the s...
{"events": [["Anandabazar Patrika", "owned by", [["ABP Group", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The aircraft involved , a Boeing 767 - 300ER , was operating as LOT Polish Airlines ' scheduled international service from Newark Liberty International Airport , New Jersey , to Warsaw . </input> Subject entity: Newark Liberty International Airport Object entity: New Jersey Your task is ...
{"events": [["Newark Liberty International Airport", "owned by", [["New Jersey", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> In 2012 , he sold his production company , So Television , to ITV for around £ 17   million . </input> Subject entity: So Television Object entity: ITV Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - Identify the re...
{"events": [["So Television", "owned by", [["ITV", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Howrah - Yesvantpur Humsafar Express Humsafar Express is a superfast express train of the Indian Railways connecting Yesvantpur Junction in Karnataka and Howrah Junction in West Bengal to . </input> Subject entity: Yesvantpur Junction Object entity: Indian Railways Your task is to identi...
{"events": [["Yesvantpur Junction", "owned by", [["Indian Railways", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The Consolidated Edison Building , also previously known as the Consolidated Gas Building , is a Neoclassical skyscraper built in 1928 . </input> Subject entity: Consolidated Gas Building Object entity: Consolidated Edison Your task is to identify the relation that holds between the subj...
{"events": [["Consolidated Gas Building", "owned by", [["Consolidated Edison", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Trains are operated by the SüdostBayernBahn network of Deutsche Bahn AG as well as by the private Bayerische Oberlandbahn railway company ( Meridian ) . </input> Subject entity: SüdostBayernBahn Object entity: Deutsche Bahn AG Your task is to identify the relation that holds between the ...
{"events": [["S\u00fcdostBayernBahn", "owned by", [["Deutsche Bahn AG", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Medway station was originally part of the New York , New Haven and Hartford Railroad , until the line was cut back to Needham Junction in July 1938 . </input> Subject entity: Medway station Object entity: New York , New Haven and Hartford Railroad Your task is to identify the relation th...
{"events": [["Medway station", "owned by", [["New York , New Haven and Hartford Railroad", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> It was bought by the York Archaeological Trust in 1987 , renamed Barley Hall , and heavily restored in a controversial project to form a museum . </input> Subject entity: Barley Hall Object entity: York Archaeological Trust Your task is to identify the relation that holds between the sub...
{"events": [["Barley Hall", "owned by", [["York Archaeological Trust", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> The Karl Renner Haus is named after the former Austrian politician and co - founder of the Friends of Nature , Karl Renner . </input> Subject entity: Karl Renner Haus Object entity: Friends of Nature Your task is to identify the relation that holds between the subject entity and the obje...
{"events": [["Karl Renner Haus", "owned by", [["Friends of Nature", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> Fraser Mansion now serves as the National Affairs office for the Church of Scientology . </input> Subject entity: Fraser Mansion Object entity: Church of Scientology Your task is to identify the relation that holds between the subject entity and the object entity. For each relation: - I...
{"events": [["Fraser Mansion", "owned by", [["Church of Scientology", "object"]], "owner of the subject"]]}
You are a relation extraction system. Your task is to identify the relation that holds between two given entities in a text. IMPORTANT:Output ONLY valid JSON. No explanations, no markdown, no extra text. Output Format (JSON only, no markdown): {"events": [[<subject span>, <relation type>, [[<object span>, "object"]], ...
Given an input text: <input> 11465/ 66 Somnath - Jabalpur Express is an Express train belonging to Indian Railways that runs between Somnath and Jabalpur Junction in India . </input> Subject entity: Somnath Object entity: Indian Railways Your task is to identify the relation that holds between the subject entity and...
{"events": [["Somnath", "owned by", [["Indian Railways", "object"]], "owner of the subject"]]}
End of preview. Expand in Data Studio

FEWREL — continual learning splits

The FEWREL corpus cut into the task streams used by the OpenED continual-learning experiments (CRE / continual relation extraction, 10 tasks per permutation, 5 permutations).

Layout

raw/{train,dev,test}.jsonl        the corpus before it is cut into tasks
perm<k>/streams.json               the label groups of permutation k, in task order
perm<k>/<task>/{train,dev,test}.jsonl
groups/                            the label groups the permutations draw from

train.jsonl of task t holds that task's labels plus a replay buffer of at most 10 examples per earlier label. test.jsonl is cumulative: every label seen up to and including task t, so a single number measures both what was learned and what was kept.

Record format

One JSON object per line, the prompt format the generative models are trained on:

{"system_prompt": "...", "user_prompt": "Given an input text: ...", "response": "{\"events\": [...]}"}

Licence

FewRel is released by THUNLP under its own terms; check before making public.

These files are a re-packaging of the original corpus: the licence of the source data applies unchanged, and this repo grants nothing beyond it.

Downloads last month
122