[{"id": "crossner_music_0", "text": "It is based on the classic song The Guns of Brixton on The Clash 's London Calling and has proven to be a success on the modern rock charts.", "requested_types": ["EVENT", "MUSIC_GENRE", "ALBUM", "AWARD", "MUSICAL_INSTRUMENT", "BAND", "SONG"], "distractors": ["MUSICAL_INSTRUMENT", "AWARD", "EVENT"], "gold": {"EVENT": [], "MUSIC_GENRE": ["rock"], "ALBUM": ["London Calling"], "AWARD": [], "MUSICAL_INSTRUMENT": [], "BAND": ["The Clash"], "SONG": ["The Guns of Brixton"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, ALBUM, AWARD, MUSICAL_INSTRUMENT, BAND, SONG\n\nText:\nIt is based on the classic song The Guns of Brixton on The Clash 's London Calling and has proven to be a success on the modern rock charts.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, ALBUM, AWARD, MUSICAL_INSTRUMENT, BAND, SONG\n\nText:\nIt is based on the classic song The Guns of Brixton on The Clash 's London Calling and has proven to be a success on the modern rock charts.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_1", "text": "As a strong advocate of animal rights, Linda lent her support to many organizations such as People for the Ethical Treatment of Animals (PETA), the Campaign to Protect Rural England, and Friends of the Earth.", "requested_types": ["BAND", "MUSICAL_ARTIST", "ORGANISATION", "COUNTRY", "MUSIC_GENRE"], "distractors": ["MUSIC_GENRE", "COUNTRY", "BAND"], "gold": {"BAND": [], "MUSICAL_ARTIST": ["Linda"], "ORGANISATION": ["People for the Ethical Treatment of Animals", "PETA", "Campaign to Protect Rural England", "Friends of the Earth"], "COUNTRY": [], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSICAL_ARTIST, ORGANISATION, COUNTRY, MUSIC_GENRE\n\nText:\nAs a strong advocate of animal rights, Linda lent her support to many organizations such as People for the Ethical Treatment of Animals (PETA), the Campaign to Protect Rural England, and Friends of the 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: BAND, MUSICAL_ARTIST, ORGANISATION, COUNTRY, MUSIC_GENRE\n\nText:\nAs a strong advocate of animal rights, Linda lent her support to many organizations such as People for the Ethical Treatment of Animals (PETA), the Campaign to Protect Rural England, and Friends of the 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_music_2", "text": "Polka is still a popular genre of folk music in many European countries and is performed by folk artists in Poland, Latvia, Lithuania, Czech Republic, Netherlands, Croatia, Slovenia, Germany, Hungary, Austria, Switzerland, Italy, Ukraine, Belarus, Russia and Slovakia.", "requested_types": ["COUNTRY", "LOCATION", "EVENT", "MUSIC_GENRE", "MISC"], "distractors": ["EVENT", "LOCATION"], "gold": {"COUNTRY": ["Poland", "Latvia", "Lithuania", "Czech Republic", "Netherlands", "Croatia", "Slovenia", "Germany", "Hungary", "Austria", "Switzerland", "Italy", "Ukraine", "Belarus", "Russia", "Slovakia"], "LOCATION": [], "EVENT": [], "MUSIC_GENRE": ["Polka", "folk music"], "MISC": ["European countries"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT, MUSIC_GENRE, MISC\n\nText:\nPolka is still a popular genre of folk music in many European countries and is performed by folk artists in Poland, Latvia, Lithuania, Czech Republic, Netherlands, Croatia, Slovenia, Germany, Hungary, Austria, Switzerland, Italy, Ukraine, Belarus, Russia and Slovakia.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT, MUSIC_GENRE, MISC\n\nText:\nPolka is still a popular genre of folk music in many European countries and is performed by folk artists in Poland, Latvia, Lithuania, Czech Republic, Netherlands, Croatia, Slovenia, Germany, Hungary, Austria, Switzerland, Italy, Ukraine, Belarus, Russia and Slovakia.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_3", "text": "Jazz accordionists from the United States include Steve Bach, Milton DeLugg, Orlando DiGirolamo, Dominic Frontiere, Guy Klucevsek, Yuri Lemeshev, Frank Marocco, John Serry Sr., Lee Tomboulian, and Art Van Damme.", "requested_types": ["AWARD", "MUSICAL_ARTIST", "MUSIC_GENRE", "COUNTRY", "ALBUM", "LOCATION"], "distractors": ["ALBUM", "LOCATION", "AWARD"], "gold": {"AWARD": [], "MUSICAL_ARTIST": ["Steve Bach", "Milton DeLugg", "Orlando DiGirolamo", "Dominic Frontiere", "Guy Klucevsek", "Yuri Lemeshev", "Frank Marocco", "John Serry Sr.", "Lee Tomboulian", "Art Van Damme"], "MUSIC_GENRE": ["Jazz"], "COUNTRY": ["United States"], "ALBUM": [], "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, MUSICAL_ARTIST, MUSIC_GENRE, COUNTRY, ALBUM, LOCATION\n\nText:\nJazz accordionists from the United States include Steve Bach, Milton DeLugg, Orlando DiGirolamo, Dominic Frontiere, Guy Klucevsek, Yuri Lemeshev, Frank Marocco, John Serry Sr., Lee Tomboulian, and Art Van Damme.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, MUSIC_GENRE, COUNTRY, ALBUM, LOCATION\n\nText:\nJazz accordionists from the United States include Steve Bach, Milton DeLugg, Orlando DiGirolamo, Dominic Frontiere, Guy Klucevsek, Yuri Lemeshev, Frank Marocco, John Serry Sr., Lee Tomboulian, and Art Van Damme.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_4", "text": "None of the singles from the album reached number one on the UK charts, but Chiquitita, Does Your Mother Know, Angeleyes (with Voulez-Vous, released as a double A-side) and I Have a Dream were all UK Top 5 hits.", "requested_types": ["EVENT", "SONG", "MISC", "COUNTRY", "ALBUM"], "distractors": ["EVENT"], "gold": {"EVENT": [], "SONG": ["Chiquitita", "Does Your Mother Know", "Angeleyes", "I Have a Dream"], "MISC": ["A-side"], "COUNTRY": ["UK"], "ALBUM": ["Voulez-Vous"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MISC, COUNTRY, ALBUM\n\nText:\nNone of the singles from the album reached number one on the UK charts, but Chiquitita, Does Your Mother Know, Angeleyes (with Voulez-Vous, released as a double A-side) and I Have a Dream were all UK Top 5 hits.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MISC, COUNTRY, ALBUM\n\nText:\nNone of the singles from the album reached number one on the UK charts, but Chiquitita, Does Your Mother Know, Angeleyes (with Voulez-Vous, released as a double A-side) and I Have a Dream were all UK Top 5 hits.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_5", "text": "Plaid went on to release several more albums, including Trainer (a compilation of early Plaid work, including the complete Mbuki Mvuki album), Double Figure, Spokes, Parts in the Post (a compilation of Plaid remixes), Greedy Baby (an audio-visual collaboration with Bob Jaroc), Scintilli, Reachy Prints and The Digging Remedy.", "requested_types": ["ALBUM", "MUSICAL_ARTIST", "MUSIC_GENRE"], "distractors": ["MUSIC_GENRE"], "gold": {"ALBUM": ["Trainer", "Mbuki Mvuki", "Double Figure", "Spokes", "Parts in the Post", "Greedy Baby", "Scintilli", "Reachy Prints", "The Digging Remedy"], "MUSICAL_ARTIST": ["Plaid", "Bob Jaroc"], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nPlaid went on to release several more albums, including Trainer (a compilation of early Plaid work, including the complete Mbuki Mvuki album), Double Figure, Spokes, Parts in the Post (a compilation of Plaid remixes), Greedy Baby (an audio-visual collaboration with Bob Jaroc), Scintilli, Reachy Prints and The Digging Remedy.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nPlaid went on to release several more albums, including Trainer (a compilation of early Plaid work, including the complete Mbuki Mvuki album), Double Figure, Spokes, Parts in the Post (a compilation of Plaid remixes), Greedy Baby (an audio-visual collaboration with Bob Jaroc), Scintilli, Reachy Prints and The Digging Remedy.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_6", "text": "The two rappers have resolved their disagreement and have collaborated on four songs: Wish You Would off Ludacris's sixth studio album, Theater of the Mind, On Top of the World off T.I. 's sixth studio album, Paper Trail and We in This Bitch off DJ Drama's album, Quality Street Music.", "requested_types": ["LOCATION", "SONG", "ALBUM", "PERSON", "MUSICAL_ARTIST"], "distractors": ["LOCATION", "PERSON"], "gold": {"LOCATION": [], "SONG": ["Wish You Would", "On Top of the World", "We in This Bitch"], "ALBUM": ["Theater of the Mind", "Paper Trail", "Quality Street Music"], "PERSON": [], "MUSICAL_ARTIST": ["Ludacris", "T.I.", "DJ Drama"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, ALBUM, PERSON, MUSICAL_ARTIST\n\nText:\nThe two rappers have resolved their disagreement and have collaborated on four songs: Wish You Would off Ludacris's sixth studio album, Theater of the Mind, On Top of the World off T.I. 's sixth studio album, Paper Trail and We in This Bitch off DJ Drama's album, Quality Street Music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, ALBUM, PERSON, MUSICAL_ARTIST\n\nText:\nThe two rappers have resolved their disagreement and have collaborated on four songs: Wish You Would off Ludacris's sixth studio album, Theater of the Mind, On Top of the World off T.I. 's sixth studio album, Paper Trail and We in This Bitch off DJ Drama's album, Quality Street Music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_7", "text": "Often termed soul blues or Southern soul, the music at the heart of this movement was given new life by the unexpected success of two particular recordings on the Jackson-based Malaco label: Z. Z. Hill 's Down Home Blues (1982) and Little Milton' s The Blues is Alright (1984).", "requested_types": ["MUSICAL_ARTIST", "SONG", "ORGANISATION", "MUSIC_GENRE", "EVENT", "BAND", "MISC"], "distractors": ["BAND", "MISC", "EVENT"], "gold": {"MUSICAL_ARTIST": ["Z. Z. Hill", "Little Milton"], "SONG": ["Down Home Blues", "The Blues is Alright"], "ORGANISATION": ["Malaco label"], "MUSIC_GENRE": ["soul blues", "Southern soul"], "EVENT": [], "BAND": [], "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: MUSICAL_ARTIST, SONG, ORGANISATION, MUSIC_GENRE, EVENT, BAND, MISC\n\nText:\nOften termed soul blues or Southern soul, the music at the heart of this movement was given new life by the unexpected success of two particular recordings on the Jackson-based Malaco label: Z. Z. Hill 's Down Home Blues (1982) and Little Milton' s The Blues is Alright (1984).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, SONG, ORGANISATION, MUSIC_GENRE, EVENT, BAND, MISC\n\nText:\nOften termed soul blues or Southern soul, the music at the heart of this movement was given new life by the unexpected success of two particular recordings on the Jackson-based Malaco label: Z. Z. Hill 's Down Home Blues (1982) and Little Milton' s The Blues is Alright (1984).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_8", "text": "The Belgium group Technotronic scored an international hit with the song Pump Up the Jam.", "requested_types": ["SONG", "EVENT", "COUNTRY", "BAND"], "distractors": ["EVENT"], "gold": {"SONG": ["Pump Up the Jam"], "EVENT": [], "COUNTRY": ["Belgium"], "BAND": ["Technotronic"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, EVENT, COUNTRY, BAND\n\nText:\nThe Belgium group Technotronic scored an international hit with the song Pump Up the Jam.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, EVENT, COUNTRY, BAND\n\nText:\nThe Belgium group Technotronic scored an international hit with the song Pump Up the Jam.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_9", "text": "Following this success, Francis recorded seven more albums of favorites between 1960 and 1964, including Connie Francis Sings Jewish Favorites, Connie Francis Sings German Favorites, and Connie Francis Sings Irish Favorites, among others.", "requested_types": ["MUSICAL_ARTIST", "ALBUM", "SONG"], "distractors": ["SONG"], "gold": {"MUSICAL_ARTIST": ["Francis"], "ALBUM": ["Connie Francis Sings Jewish Favorites", "Connie Francis Sings German Favorites", "Connie Francis Sings Irish Favorites"], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ALBUM, SONG\n\nText:\nFollowing this success, Francis recorded seven more albums of favorites between 1960 and 1964, including Connie Francis Sings Jewish Favorites, Connie Francis Sings German Favorites, and Connie Francis Sings Irish Favorites, 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: MUSICAL_ARTIST, ALBUM, SONG\n\nText:\nFollowing this success, Francis recorded seven more albums of favorites between 1960 and 1964, including Connie Francis Sings Jewish Favorites, Connie Francis Sings German Favorites, and Connie Francis Sings Irish Favorites, 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_music_10", "text": "With Fredriksson's continued treatment and recuperation he made three more albums, Son of a Plumber (2005), En händig man (A handy man, 2007) and Party Crasher (2008).", "requested_types": ["COUNTRY", "SONG", "MUSICAL_ARTIST", "ALBUM", "MISC"], "distractors": ["COUNTRY", "MISC", "SONG"], "gold": {"COUNTRY": [], "SONG": [], "MUSICAL_ARTIST": ["Fredriksson"], "ALBUM": ["Son of a Plumber", "En händig man", "A handy man", "Party Crasher"], "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: COUNTRY, SONG, MUSICAL_ARTIST, ALBUM, MISC\n\nText:\nWith Fredriksson's continued treatment and recuperation he made three more albums, Son of a Plumber (2005), En händig man (A handy man, 2007) and Party Crasher (2008).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST, ALBUM, MISC\n\nText:\nWith Fredriksson's continued treatment and recuperation he made three more albums, Son of a Plumber (2005), En händig man (A handy man, 2007) and Party Crasher (2008).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_11", "text": "In April 2019, Phair announced via her Instagram that she had been working on new studio material with Brad Wood, who produced Exile in Guyville, Whip-Smart, and parts of Whitechocolatespaceegg.", "requested_types": ["ALBUM", "MISC", "MUSIC_GENRE", "MUSICAL_ARTIST", "COUNTRY"], "distractors": ["MUSIC_GENRE", "COUNTRY"], "gold": {"ALBUM": ["Exile in Guyville", "Whip-Smart", "Whitechocolatespaceegg"], "MISC": ["Instagram"], "MUSIC_GENRE": [], "MUSICAL_ARTIST": ["Phair", "Brad Wood"], "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: ALBUM, MISC, MUSIC_GENRE, MUSICAL_ARTIST, COUNTRY\n\nText:\nIn April 2019, Phair announced via her Instagram that she had been working on new studio material with Brad Wood, who produced Exile in Guyville, Whip-Smart, and parts of Whitechocolatespaceegg.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MISC, MUSIC_GENRE, MUSICAL_ARTIST, COUNTRY\n\nText:\nIn April 2019, Phair announced via her Instagram that she had been working on new studio material with Brad Wood, who produced Exile in Guyville, Whip-Smart, and parts of Whitechocolatespaceegg.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_12", "text": "Annie Hall (1977) won four Academy Awards, including Academy Award for Best Picture, Academy Award for Best Actress for Diane Keaton, Academy Award for Best Original Screenplay and Academy Award for Best Director for Woody Allen.", "requested_types": ["AWARD", "MISC", "ORGANISATION", "MUSIC_GENRE", "PERSON"], "distractors": ["MUSIC_GENRE", "ORGANISATION"], "gold": {"AWARD": ["Academy Awards", "Academy Award for Best Picture", "Academy Award for Best Actress", "Academy Award for Best Original Screenplay", "Academy Award for Best Director"], "MISC": ["Annie Hall"], "ORGANISATION": [], "MUSIC_GENRE": [], "PERSON": ["Diane Keaton", "Woody Allen"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, PERSON\n\nText:\nAnnie Hall (1977) won four Academy Awards, including Academy Award for Best Picture, Academy Award for Best Actress for Diane Keaton, Academy Award for Best Original Screenplay and Academy Award for Best Director for Woody Allen.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, PERSON\n\nText:\nAnnie Hall (1977) won four Academy Awards, including Academy Award for Best Picture, Academy Award for Best Actress for Diane Keaton, Academy Award for Best Original Screenplay and Academy Award for Best Director for Woody Allen.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_13", "text": "Also in 1993, the band contributed the track It's the New Style (with DJ Hurricane) to the AIDS benefit album No Alternative, produced by the Red Hot Organization.", "requested_types": ["ORGANISATION", "MUSICAL_ARTIST", "PERSON", "MISC", "ALBUM", "SONG"], "distractors": ["PERSON"], "gold": {"ORGANISATION": ["Red Hot Organization"], "MUSICAL_ARTIST": ["DJ Hurricane"], "PERSON": [], "MISC": ["AIDS benefit album"], "ALBUM": ["No Alternative"], "SONG": ["the New Style"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, PERSON, MISC, ALBUM, SONG\n\nText:\nAlso in 1993, the band contributed the track It's the New Style (with DJ Hurricane) to the AIDS benefit album No Alternative, produced by the Red Hot Organization.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, PERSON, MISC, ALBUM, SONG\n\nText:\nAlso in 1993, the band contributed the track It's the New Style (with DJ Hurricane) to the AIDS benefit album No Alternative, produced by the Red Hot Organization.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_14", "text": "Often cited as a pop icon, Timberlake is the recipient of numerous awards and accolades, including ten Grammy Award s, four Emmy Award s, three Brit Awards, nine Billboard Music Awards, the Contemporary Icon Award by the Songwriters Hall of Fame, an honorary Doctor of Music degree from the Berklee College of Music, and the Michael Jackson Video Vanguard Award.", "requested_types": ["ORGANISATION", "MUSIC_GENRE", "SONG", "ALBUM", "AWARD", "MUSICAL_ARTIST"], "distractors": ["MUSIC_GENRE", "ALBUM", "SONG"], "gold": {"ORGANISATION": ["Songwriters Hall of Fame", "Berklee College of Music"], "MUSIC_GENRE": [], "SONG": [], "ALBUM": [], "AWARD": ["Grammy Award", "Emmy Award", "Brit Awards", "Billboard Music Awards", "Contemporary Icon Award", "honorary Doctor of Music degree", "Michael Jackson Video Vanguard Award"], "MUSICAL_ARTIST": ["Timberlake"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, SONG, ALBUM, AWARD, MUSICAL_ARTIST\n\nText:\nOften cited as a pop icon, Timberlake is the recipient of numerous awards and accolades, including ten Grammy Award s, four Emmy Award s, three Brit Awards, nine Billboard Music Awards, the Contemporary Icon Award by the Songwriters Hall of Fame, an honorary Doctor of Music degree from the Berklee College of Music, and the Michael Jackson Video Vanguard 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, MUSIC_GENRE, SONG, ALBUM, AWARD, MUSICAL_ARTIST\n\nText:\nOften cited as a pop icon, Timberlake is the recipient of numerous awards and accolades, including ten Grammy Award s, four Emmy Award s, three Brit Awards, nine Billboard Music Awards, the Contemporary Icon Award by the Songwriters Hall of Fame, an honorary Doctor of Music degree from the Berklee College of Music, and the Michael Jackson Video Vanguard 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_music_15", "text": "Several notable albums have been released, in particular the Morris On series, which consists of Morris On, Son of Morris On, Grandson of Morris On, Great Grandson of Morris On, Morris on the Road, and Mother of all Morris.", "requested_types": ["ALBUM", "ORGANISATION", "PERSON", "BAND"], "distractors": ["BAND", "ORGANISATION", "PERSON"], "gold": {"ALBUM": ["Morris On", "Son of Morris On", "Grandson of Morris On", "Great Grandson of Morris On", "Morris on the Road", "Mother of all Morris"], "ORGANISATION": [], "PERSON": [], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, ORGANISATION, PERSON, BAND\n\nText:\nSeveral notable albums have been released, in particular the Morris On series, which consists of Morris On, Son of Morris On, Grandson of Morris On, Great Grandson of Morris On, Morris on the Road, and Mother of all 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: ALBUM, ORGANISATION, PERSON, BAND\n\nText:\nSeveral notable albums have been released, in particular the Morris On series, which consists of Morris On, Son of Morris On, Grandson of Morris On, Great Grandson of Morris On, Morris on the Road, and Mother of all 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_music_16", "text": "The film won Academy Awards for Academy Award for Best Costume Design, Academy Award for Best Makeup and Hairstyling and Academy Award for Best Sound Editing.", "requested_types": ["AWARD", "MISC", "MUSIC_GENRE", "COUNTRY"], "distractors": ["MISC", "MUSIC_GENRE", "COUNTRY"], "gold": {"AWARD": ["Academy Awards", "Academy Award for Best Costume Design", "Academy Award for Best Makeup and Hairstyling", "Academy Award for Best Sound Editing"], "MISC": [], "MUSIC_GENRE": [], "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, MISC, MUSIC_GENRE, COUNTRY\n\nText:\nThe film won Academy Awards for Academy Award for Best Costume Design, Academy Award for Best Makeup and Hairstyling and Academy Award for Best Sound 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: AWARD, MISC, MUSIC_GENRE, COUNTRY\n\nText:\nThe film won Academy Awards for Academy Award for Best Costume Design, Academy Award for Best Makeup and Hairstyling and Academy Award for Best Sound 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_music_17", "text": "In the early 2010s, there was somewhat of a resurgence of boy band popularity in countries where the trend had not maintained, with the emergence of new boy bands like Big Time Rush, The Wanted, and One Direction and the formation of supergroup NKOTBSB which comprised members of New Kids on the Block and Backstreet Boys.", "requested_types": ["BAND", "ALBUM"], "distractors": ["ALBUM"], "gold": {"BAND": ["Big Time Rush", "The Wanted", "One Direction", "NKOTBSB", "New Kids on the Block", "Backstreet Boys"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, ALBUM\n\nText:\nIn the early 2010s, there was somewhat of a resurgence of boy band popularity in countries where the trend had not maintained, with the emergence of new boy bands like Big Time Rush, The Wanted, and One Direction and the formation of supergroup NKOTBSB which comprised members of New Kids on the Block and Backstreet Boys.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, ALBUM\n\nText:\nIn the early 2010s, there was somewhat of a resurgence of boy band popularity in countries where the trend had not maintained, with the emergence of new boy bands like Big Time Rush, The Wanted, and One Direction and the formation of supergroup NKOTBSB which comprised members of New Kids on the Block and Backstreet Boys.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_18", "text": "The first wave of black metal emerged in Europe in the early and mid-1980s, led by the United Kingdom's Venom, Denmark's Mercyful Fate, Switzerland's Hellhammer and Celtic Frost, and Sweden's Bathory.", "requested_types": ["MUSIC_GENRE", "MUSICAL_ARTIST", "COUNTRY", "BAND", "ALBUM", "ORGANISATION", "LOCATION"], "distractors": ["ORGANISATION", "MUSICAL_ARTIST", "ALBUM"], "gold": {"MUSIC_GENRE": ["black metal"], "MUSICAL_ARTIST": [], "COUNTRY": ["United Kingdom", "Denmark", "Switzerland", "Sweden"], "BAND": ["Venom", "Mercyful Fate", "Hellhammer", "Celtic Frost", "Bathory"], "ALBUM": [], "ORGANISATION": [], "LOCATION": ["Europe"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, COUNTRY, BAND, ALBUM, ORGANISATION, LOCATION\n\nText:\nThe first wave of black metal emerged in Europe in the early and mid-1980s, led by the United Kingdom's Venom, Denmark's Mercyful Fate, Switzerland's Hellhammer and Celtic Frost, and Sweden's Bathory.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, COUNTRY, BAND, ALBUM, ORGANISATION, LOCATION\n\nText:\nThe first wave of black metal emerged in Europe in the early and mid-1980s, led by the United Kingdom's Venom, Denmark's Mercyful Fate, Switzerland's Hellhammer and Celtic Frost, and Sweden's Bathory.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_19", "text": "Beginning in 1999, Righteous Babe Records began releasing albums by other artists including Sara Lee, Sekou Sundiata, Michael Meldrum, Arto Lindsay, Bitch and Animal, That One Guy, Utah Phillips, Hamell on Trial, Andrew Bird, Kurt Swinghammer, Buddy Wakefield, Anaïs Mitchell and Nona Hendryx.", "requested_types": ["PERSON", "MUSICAL_ARTIST", "MUSIC_GENRE", "ORGANISATION", "BAND"], "distractors": ["MUSIC_GENRE", "PERSON"], "gold": {"PERSON": [], "MUSICAL_ARTIST": ["Sara Lee", "Sekou Sundiata", "Michael Meldrum", "Arto Lindsay", "That One Guy", "Utah Phillips", "Andrew Bird", "Kurt Swinghammer", "Buddy Wakefield", "Anaïs Mitchell", "Nona Hendryx"], "MUSIC_GENRE": [], "ORGANISATION": ["Righteous Babe Records"], "BAND": ["Bitch and Animal", "Hamell on Trial"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, MUSIC_GENRE, ORGANISATION, BAND\n\nText:\nBeginning in 1999, Righteous Babe Records began releasing albums by other artists including Sara Lee, Sekou Sundiata, Michael Meldrum, Arto Lindsay, Bitch and Animal, That One Guy, Utah Phillips, Hamell on Trial, Andrew Bird, Kurt Swinghammer, Buddy Wakefield, Anaïs Mitchell and Nona Hendryx.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, MUSIC_GENRE, ORGANISATION, BAND\n\nText:\nBeginning in 1999, Righteous Babe Records began releasing albums by other artists including Sara Lee, Sekou Sundiata, Michael Meldrum, Arto Lindsay, Bitch and Animal, That One Guy, Utah Phillips, Hamell on Trial, Andrew Bird, Kurt Swinghammer, Buddy Wakefield, Anaïs Mitchell and Nona Hendryx.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_20", "text": "Starting in the early 1950s, and during the mid-1960s, Western singer-songwriters such as Michael Martin Murphey and Marty Robbins rose in prominence as did others, throughout Western music traditions, like New Mexico music 's Al Hurricane.", "requested_types": ["ALBUM", "MUSICAL_ARTIST", "MUSIC_GENRE", "MISC"], "distractors": ["ALBUM", "MISC"], "gold": {"ALBUM": [], "MUSICAL_ARTIST": ["Michael Martin Murphey", "Marty Robbins", "Al Hurricane"], "MUSIC_GENRE": ["Western music", "New Mexico music"], "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: ALBUM, MUSICAL_ARTIST, MUSIC_GENRE, MISC\n\nText:\nStarting in the early 1950s, and during the mid-1960s, Western singer-songwriters such as Michael Martin Murphey and Marty Robbins rose in prominence as did others, throughout Western music traditions, like New Mexico music 's Al Hurricane.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_ARTIST, MUSIC_GENRE, MISC\n\nText:\nStarting in the early 1950s, and during the mid-1960s, Western singer-songwriters such as Michael Martin Murphey and Marty Robbins rose in prominence as did others, throughout Western music traditions, like New Mexico music 's Al Hurricane.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_21", "text": "Under the direction of producers such as Chet Atkins, Bill Porter, Paul Cohen, Owen Bradley, Bob Ferguson, and later Billy Sherrill, the sound brought country music to a diverse audience and helped revive country as it emerged from a commercially fallow period.", "requested_types": ["MUSICAL_ARTIST", "BAND", "MUSICAL_INSTRUMENT", "MUSIC_GENRE"], "distractors": ["MUSICAL_INSTRUMENT", "BAND"], "gold": {"MUSICAL_ARTIST": ["Chet Atkins", "Bill Porter", "Paul Cohen", "Owen Bradley", "Bob Ferguson", "Billy Sherrill"], "BAND": [], "MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": ["country music", "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: MUSICAL_ARTIST, BAND, MUSICAL_INSTRUMENT, MUSIC_GENRE\n\nText:\nUnder the direction of producers such as Chet Atkins, Bill Porter, Paul Cohen, Owen Bradley, Bob Ferguson, and later Billy Sherrill, the sound brought country music to a diverse audience and helped revive country as it emerged from a commercially fallow period.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, BAND, MUSICAL_INSTRUMENT, MUSIC_GENRE\n\nText:\nUnder the direction of producers such as Chet Atkins, Bill Porter, Paul Cohen, Owen Bradley, Bob Ferguson, and later Billy Sherrill, the sound brought country music to a diverse audience and helped revive country as it emerged from a commercially fallow period.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_22", "text": "The conductors with whom she has worked include Pierre Boulez, Sir Simon Rattle, Sir Colin Davis, Michael Tilson Thomas, and Valery Gergiev, and she has appeared in many of the world's greatest venues, including the Wigmore Hall, Southbank Centre and the Barbican Centre in London, Sydney Opera House, New York's Lincoln Center, Leipzig Gewandhaus, the Concertgebouw in Amsterdam and the Mozarteum in Salzburg.", "requested_types": ["MUSIC_GENRE", "LOCATION", "BAND", "ALBUM", "MUSICAL_ARTIST"], "distractors": ["BAND", "ALBUM", "MUSIC_GENRE"], "gold": {"MUSIC_GENRE": [], "LOCATION": ["Wigmore Hall", "Southbank Centre", "Barbican Centre", "London", "Sydney Opera House", "New York", "Lincoln Center", "Leipzig Gewandhaus", "Concertgebouw", "Amsterdam", "Mozarteum", "Salzburg"], "BAND": [], "ALBUM": [], "MUSICAL_ARTIST": ["Pierre Boulez", "Simon Rattle", "Colin Davis", "Michael Tilson Thomas", "Valery Gergiev"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, LOCATION, BAND, ALBUM, MUSICAL_ARTIST\n\nText:\nThe conductors with whom she has worked include Pierre Boulez, Sir Simon Rattle, Sir Colin Davis, Michael Tilson Thomas, and Valery Gergiev, and she has appeared in many of the world's greatest venues, including the Wigmore Hall, Southbank Centre and the Barbican Centre in London, Sydney Opera House, New York's Lincoln Center, Leipzig Gewandhaus, the Concertgebouw in Amsterdam and the Mozarteum in Salzburg.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, LOCATION, BAND, ALBUM, MUSICAL_ARTIST\n\nText:\nThe conductors with whom she has worked include Pierre Boulez, Sir Simon Rattle, Sir Colin Davis, Michael Tilson Thomas, and Valery Gergiev, and she has appeared in many of the world's greatest venues, including the Wigmore Hall, Southbank Centre and the Barbican Centre in London, Sydney Opera House, New York's Lincoln Center, Leipzig Gewandhaus, the Concertgebouw in Amsterdam and the Mozarteum in Salzburg.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_23", "text": "In 1993, Grohl was recruited to help recreate the music of The Beatles' early years for the movie Backbeat; he played drums in an all-star lineup that included Greg Dull of the Afghan Whigs, indie producer Don Fleming, Mike Mills of R.E.M., Thurston Moore of Sonic Youth, and Dave Pirner of Soul Asylum.", "requested_types": ["MUSIC_GENRE", "MISC", "MUSICAL_ARTIST", "BAND", "ALBUM", "SONG", "MUSICAL_INSTRUMENT"], "distractors": ["SONG", "ALBUM"], "gold": {"MUSIC_GENRE": ["indie"], "MISC": ["Backbeat"], "MUSICAL_ARTIST": ["Grohl", "Greg Dull", "Don Fleming", "Mike Mills", "Thurston Moore", "Dave Pirner"], "BAND": ["The Beatles", "the Afghan Whigs", "R.E.M.", "Sonic Youth", "Soul Asylum"], "ALBUM": [], "SONG": [], "MUSICAL_INSTRUMENT": ["drums"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MISC, MUSICAL_ARTIST, BAND, ALBUM, SONG, MUSICAL_INSTRUMENT\n\nText:\nIn 1993, Grohl was recruited to help recreate the music of The Beatles' early years for the movie Backbeat; he played drums in an all-star lineup that included Greg Dull of the Afghan Whigs, indie producer Don Fleming, Mike Mills of R.E.M., Thurston Moore of Sonic Youth, and Dave Pirner of Soul Asylum.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MISC, MUSICAL_ARTIST, BAND, ALBUM, SONG, MUSICAL_INSTRUMENT\n\nText:\nIn 1993, Grohl was recruited to help recreate the music of The Beatles' early years for the movie Backbeat; he played drums in an all-star lineup that included Greg Dull of the Afghan Whigs, indie producer Don Fleming, Mike Mills of R.E.M., Thurston Moore of Sonic Youth, and Dave Pirner of Soul Asylum.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_24", "text": "The venues they visited were (in chronological order), Southampton Mayflower Theatre, Leeds Grand Theatre, Southend Cliffs Pavilion, Bristol Hippodrome, Bournemouth International Centre, and Milton Keynes Theatre.", "requested_types": ["PERSON", "SONG", "LOCATION"], "distractors": ["SONG", "PERSON"], "gold": {"PERSON": [], "SONG": [], "LOCATION": ["Southampton Mayflower Theatre", "Leeds Grand Theatre", "Southend Cliffs Pavilion", "Bristol Hippodrome", "Bournemouth International Centre", "Milton Keynes Theatre"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, LOCATION\n\nText:\nThe venues they visited were (in chronological order), Southampton Mayflower Theatre, Leeds Grand Theatre, Southend Cliffs Pavilion, Bristol Hippodrome, Bournemouth International Centre, and Milton Keynes Theatre.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, LOCATION\n\nText:\nThe venues they visited were (in chronological order), Southampton Mayflower Theatre, Leeds Grand Theatre, Southend Cliffs Pavilion, Bristol Hippodrome, Bournemouth International Centre, and Milton Keynes Theatre.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_25", "text": "The Convention Centre hosted preliminary rounds of the men's basketball competition at the Commonwealth Games, which was held on Queensland's Gold Coast in April 2018 Commonwealth Games.", "requested_types": ["EVENT", "MUSICAL_INSTRUMENT", "LOCATION", "AWARD", "ORGANISATION"], "distractors": ["MUSICAL_INSTRUMENT", "AWARD", "ORGANISATION"], "gold": {"EVENT": ["men's basketball competition", "Commonwealth Games", "2018 Commonwealth Games"], "MUSICAL_INSTRUMENT": [], "LOCATION": ["Convention Centre", "Queensland's Gold Coast"], "AWARD": [], "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, MUSICAL_INSTRUMENT, LOCATION, AWARD, ORGANISATION\n\nText:\nThe Convention Centre hosted preliminary rounds of the men's basketball competition at the Commonwealth Games, which was held on Queensland's Gold Coast in April 2018 Commonwealth 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, MUSICAL_INSTRUMENT, LOCATION, AWARD, ORGANISATION\n\nText:\nThe Convention Centre hosted preliminary rounds of the men's basketball competition at the Commonwealth Games, which was held on Queensland's Gold Coast in April 2018 Commonwealth 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_music_26", "text": "After Gulf Winds (1976), an album of entirely self-composed songs and From Every Stage (1976), a live album that had Baez performing songs from every stage of her career, Baez again parted ways with a record label when she moved to CBS Records for Blowin 'Away (1977) and Honest Lullaby (1979).", "requested_types": ["ALBUM", "BAND", "MUSICAL_ARTIST", "ORGANISATION"], "distractors": ["BAND"], "gold": {"ALBUM": ["Gulf Winds", "From Every Stage", "Blowin 'Away", "Honest Lullaby"], "BAND": [], "MUSICAL_ARTIST": ["Baez"], "ORGANISATION": ["CBS Records"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, BAND, MUSICAL_ARTIST, ORGANISATION\n\nText:\nAfter Gulf Winds (1976), an album of entirely self-composed songs and From Every Stage (1976), a live album that had Baez performing songs from every stage of her career, Baez again parted ways with a record label when she moved to CBS Records for Blowin 'Away (1977) and Honest Lullaby (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: ALBUM, BAND, MUSICAL_ARTIST, ORGANISATION\n\nText:\nAfter Gulf Winds (1976), an album of entirely self-composed songs and From Every Stage (1976), a live album that had Baez performing songs from every stage of her career, Baez again parted ways with a record label when she moved to CBS Records for Blowin 'Away (1977) and Honest Lullaby (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_music_27", "text": "He played on the live albums that would follow the release of Bitches Brew, taken from concerts at the Fillmore East in New York and Fillmore West in San Francisco.", "requested_types": ["EVENT", "LOCATION", "ALBUM"], "distractors": ["EVENT"], "gold": {"EVENT": [], "LOCATION": ["Fillmore East", "New York", "Fillmore West", "San Francisco"], "ALBUM": ["Bitches Brew"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nHe played on the live albums that would follow the release of Bitches Brew, taken from concerts at the Fillmore East in New York and Fillmore West in San Francisco.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nHe played on the live albums that would follow the release of Bitches Brew, taken from concerts at the Fillmore East in New York and Fillmore West in San Francisco.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_28", "text": "Over the years, the festival has taken place in numerous venues, including the Angel Orensanz Center for the Arts, the St. Nicholas of Myra church basement, the New Age Cabaret (formerly known as the Electric Circus), the Knitting Factory, St. Patrick's youth center, CBGB, Clemente Soto Vélez Cultural Center, the Abrons Arts Center, Roulette Intermedium and Judson Memorial Church.", "requested_types": ["AWARD", "LOCATION", "EVENT"], "distractors": ["AWARD", "EVENT"], "gold": {"AWARD": [], "LOCATION": ["Angel Orensanz Center", "St. Nicholas of Myra church basement", "New Age Cabaret", "Electric Circus", "Knitting Factory", "St. Patrick's youth center", "CBGB", "Clemente Soto Vélez Cultural Center", "Abrons Arts Center", "Roulette Intermedium", "Judson Memorial Church"], "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, LOCATION, EVENT\n\nText:\nOver the years, the festival has taken place in numerous venues, including the Angel Orensanz Center for the Arts, the St. Nicholas of Myra church basement, the New Age Cabaret (formerly known as the Electric Circus), the Knitting Factory, St. Patrick's youth center, CBGB, Clemente Soto Vélez Cultural Center, the Abrons Arts Center, Roulette Intermedium and Judson Memorial Church.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, EVENT\n\nText:\nOver the years, the festival has taken place in numerous venues, including the Angel Orensanz Center for the Arts, the St. Nicholas of Myra church basement, the New Age Cabaret (formerly known as the Electric Circus), the Knitting Factory, St. Patrick's youth center, CBGB, Clemente Soto Vélez Cultural Center, the Abrons Arts Center, Roulette Intermedium and Judson Memorial Church.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_29", "text": "Her charting singles include Crucify, Silent All These Years, God, Cornflake Girl, Caught a Lite Sneeze, Professional Widow, Spark, 1000 Oceans, Flavor and A Sorta Fairytale, her most commercially successful single in the U.S. to date.", "requested_types": ["SONG", "COUNTRY", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_ARTIST"], "gold": {"SONG": ["Crucify", "Silent All These Years", "God", "Cornflake Girl", "Caught a Lite Sneeze", "Professional Widow", "Spark", "1000 Oceans", "Flavor", "A Sorta Fairytale"], "COUNTRY": ["U.S."], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, COUNTRY, MUSICAL_ARTIST\n\nText:\nHer charting singles include Crucify, Silent All These Years, God, Cornflake Girl, Caught a Lite Sneeze, Professional Widow, Spark, 1000 Oceans, Flavor and A Sorta Fairytale, her most commercially successful single in the U.S. to date.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, COUNTRY, MUSICAL_ARTIST\n\nText:\nHer charting singles include Crucify, Silent All These Years, God, Cornflake Girl, Caught a Lite Sneeze, Professional Widow, Spark, 1000 Oceans, Flavor and A Sorta Fairytale, her most commercially successful single in the U.S. to date.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_30", "text": "In November 2011, three of the band's albums (The Power Cosmic, Atlantis Ascendant and The Chthonic Chronicles) were reissued as limited edition digipaks by Nuclear Blast's affiliate label Metal Mind Productions.", "requested_types": ["ALBUM", "AWARD", "ORGANISATION", "MUSICAL_INSTRUMENT", "SONG"], "distractors": ["SONG", "AWARD", "MUSICAL_INSTRUMENT"], "gold": {"ALBUM": ["The Power Cosmic", "Atlantis Ascendant", "The Chthonic Chronicles"], "AWARD": [], "ORGANISATION": ["Nuclear Blast", "Metal Mind Productions"], "MUSICAL_INSTRUMENT": [], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, AWARD, ORGANISATION, MUSICAL_INSTRUMENT, SONG\n\nText:\nIn November 2011, three of the band's albums (The Power Cosmic, Atlantis Ascendant and The Chthonic Chronicles) were reissued as limited edition digipaks by Nuclear Blast's affiliate label Metal Mind Productions.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, AWARD, ORGANISATION, MUSICAL_INSTRUMENT, SONG\n\nText:\nIn November 2011, three of the band's albums (The Power Cosmic, Atlantis Ascendant and The Chthonic Chronicles) were reissued as limited edition digipaks by Nuclear Blast's affiliate label Metal Mind Productions.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_31", "text": "The style of British blues developed in the UK, when bands such as the The Animals, Fleetwood Mac, John Mayall & the Bluesbreakers, the The Rolling Stones, the The Yardbirds, the supergroup Cream and the Irish musician Rory Gallagher performed classic blues songs from the Delta blues or Chicago blues traditions.", "requested_types": ["SONG", "MUSIC_GENRE", "EVENT", "MISC", "AWARD", "COUNTRY", "BAND", "MUSICAL_ARTIST"], "distractors": ["SONG", "AWARD", "EVENT"], "gold": {"SONG": [], "MUSIC_GENRE": ["British blues", "blues", "Delta blues", "Chicago blues"], "EVENT": [], "MISC": ["Irish"], "AWARD": [], "COUNTRY": ["UK"], "BAND": ["The Animals", "Fleetwood Mac", "John Mayall & the Bluesbreakers", "The Rolling Stones", "The Yardbirds", "supergroup Cream"], "MUSICAL_ARTIST": ["Rory Gallagher"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE, EVENT, MISC, AWARD, COUNTRY, BAND, MUSICAL_ARTIST\n\nText:\nThe style of British blues developed in the UK, when bands such as the The Animals, Fleetwood Mac, John Mayall & the Bluesbreakers, the The Rolling Stones, the The Yardbirds, the supergroup Cream and the Irish musician Rory Gallagher performed classic blues songs from the Delta blues or Chicago blues traditions.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE, EVENT, MISC, AWARD, COUNTRY, BAND, MUSICAL_ARTIST\n\nText:\nThe style of British blues developed in the UK, when bands such as the The Animals, Fleetwood Mac, John Mayall & the Bluesbreakers, the The Rolling Stones, the The Yardbirds, the supergroup Cream and the Irish musician Rory Gallagher performed classic blues songs from the Delta blues or Chicago blues traditions.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_32", "text": "During the 1970s she returned to traditional jazz and recorded Judith Durham and The Hottest Band in Town and Judith Durham and The Hottest Band in Town Volume 2 and in 1978, The Hot Jazz Duo.", "requested_types": ["MUSIC_GENRE", "BAND", "ALBUM", "EVENT"], "distractors": ["EVENT", "BAND"], "gold": {"MUSIC_GENRE": ["traditional jazz"], "BAND": [], "ALBUM": ["Judith Durham and The Hottest Band in Town", "Judith Durham and The Hottest Band in Town Volume 2", "The Hot Jazz Duo"], "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: MUSIC_GENRE, BAND, ALBUM, EVENT\n\nText:\nDuring the 1970s she returned to traditional jazz and recorded Judith Durham and The Hottest Band in Town and Judith Durham and The Hottest Band in Town Volume 2 and in 1978, The Hot Jazz Duo.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, BAND, ALBUM, EVENT\n\nText:\nDuring the 1970s she returned to traditional jazz and recorded Judith Durham and The Hottest Band in Town and Judith Durham and The Hottest Band in Town Volume 2 and in 1978, The Hot Jazz Duo.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_33", "text": "She later released a CD on Righteous Babe of the concert Til We Outnumber Em featuring artists such as DiFranco, Billy Bragg, Ramblin 'Jack Elliott, Arlo Guthrie, Indigo Girls, Dave Pirner, Tim Robbins, and Bruce Springsteen with 100 percent of proceeds going to the Woody Guthrie Foundation and Archives and the Rock and Roll Hall of Fame Museum educational department.", "requested_types": ["ALBUM", "MUSIC_GENRE", "PERSON", "ORGANISATION", "MISC", "BAND", "SONG", "MUSICAL_ARTIST"], "distractors": ["MUSIC_GENRE", "SONG", "PERSON"], "gold": {"ALBUM": ["Til We Outnumber Em"], "MUSIC_GENRE": [], "PERSON": [], "ORGANISATION": ["Righteous Babe", "Woody Guthrie Foundation", "Archives", "Rock and Roll Hall of Fame Museum"], "MISC": ["CD"], "BAND": ["Indigo Girls"], "SONG": [], "MUSICAL_ARTIST": ["DiFranco", "Billy Bragg", "Ramblin 'Jack Elliott", "Arlo Guthrie", "Dave Pirner", "Tim Robbins", "Bruce Springsteen"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSIC_GENRE, PERSON, ORGANISATION, MISC, BAND, SONG, MUSICAL_ARTIST\n\nText:\nShe later released a CD on Righteous Babe of the concert Til We Outnumber Em featuring artists such as DiFranco, Billy Bragg, Ramblin 'Jack Elliott, Arlo Guthrie, Indigo Girls, Dave Pirner, Tim Robbins, and Bruce Springsteen with 100 percent of proceeds going to the Woody Guthrie Foundation and Archives and the Rock and Roll Hall of Fame Museum educational department.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSIC_GENRE, PERSON, ORGANISATION, MISC, BAND, SONG, MUSICAL_ARTIST\n\nText:\nShe later released a CD on Righteous Babe of the concert Til We Outnumber Em featuring artists such as DiFranco, Billy Bragg, Ramblin 'Jack Elliott, Arlo Guthrie, Indigo Girls, Dave Pirner, Tim Robbins, and Bruce Springsteen with 100 percent of proceeds going to the Woody Guthrie Foundation and Archives and the Rock and Roll Hall of Fame Museum educational department.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_34", "text": "Other artists featured on the show include Michael Jackson, Barry White, Al Green, Tina Turner, Macy Gray, Gloria Gaynor, Chayanne, Barry Manilow, Anastacia, Elton John, Sting and Mariah Carey.", "requested_types": ["ORGANISATION", "MUSICAL_ARTIST"], "distractors": ["ORGANISATION"], "gold": {"ORGANISATION": [], "MUSICAL_ARTIST": ["Michael Jackson", "Barry White", "Al Green", "Tina Turner", "Macy Gray", "Gloria Gaynor", "Chayanne", "Barry Manilow", "Anastacia", "Elton John", "Sting", "Mariah Carey"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST\n\nText:\nOther artists featured on the show include Michael Jackson, Barry White, Al Green, Tina Turner, Macy Gray, Gloria Gaynor, Chayanne, Barry Manilow, Anastacia, Elton John, Sting and Mariah Carey.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST\n\nText:\nOther artists featured on the show include Michael Jackson, Barry White, Al Green, Tina Turner, Macy Gray, Gloria Gaynor, Chayanne, Barry Manilow, Anastacia, Elton John, Sting and Mariah Carey.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_35", "text": "The festival presents concerts in Dixieland, swing, Jazz fusion, blues, gospel, funk, soul and drum and bass.", "requested_types": ["COUNTRY", "MUSIC_GENRE", "LOCATION", "MISC"], "distractors": ["MISC", "LOCATION", "COUNTRY"], "gold": {"COUNTRY": [], "MUSIC_GENRE": ["Dixieland", "swing", "Jazz fusion", "blues", "gospel", "funk", "soul", "drum and bass"], "LOCATION": [], "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: COUNTRY, MUSIC_GENRE, LOCATION, MISC\n\nText:\nThe festival presents concerts in Dixieland, swing, Jazz fusion, blues, gospel, funk, soul and drum and bass.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, LOCATION, MISC\n\nText:\nThe festival presents concerts in Dixieland, swing, Jazz fusion, blues, gospel, funk, soul and drum and bass.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_36", "text": "Duke turned 65 in the spring of 1964 but showed no signs of slowing down as he continued to make vital and innovative recordings, including The Far East Suite (1966), New Orleans Suite (1970), Latin American Suite (1972) and The Afro-Eurasian Eclipse (1971), much of it inspired by his world tours.", "requested_types": ["MISC", "AWARD", "MUSICAL_ARTIST", "ALBUM"], "distractors": ["AWARD", "MISC"], "gold": {"MISC": [], "AWARD": [], "MUSICAL_ARTIST": ["Duke"], "ALBUM": ["The Far East Suite", "New Orleans Suite", "Latin American Suite", "The Afro-Eurasian Eclipse"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ALBUM\n\nText:\nDuke turned 65 in the spring of 1964 but showed no signs of slowing down as he continued to make vital and innovative recordings, including The Far East Suite (1966), New Orleans Suite (1970), Latin American Suite (1972) and The Afro-Eurasian Eclipse (1971), much of it inspired by his world tours.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ALBUM\n\nText:\nDuke turned 65 in the spring of 1964 but showed no signs of slowing down as he continued to make vital and innovative recordings, including The Far East Suite (1966), New Orleans Suite (1970), Latin American Suite (1972) and The Afro-Eurasian Eclipse (1971), much of it inspired by his world tours.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_37", "text": "Prince formed the Time, originally conceived as an opening act for him and based on his Minneapolis sound, a hybrid mixture of funk, R & B, Rock music, Pop music & amp; New wave music.", "requested_types": ["MUSIC_GENRE", "BAND", "MUSICAL_ARTIST", "EVENT", "COUNTRY", "SONG"], "distractors": ["SONG", "COUNTRY", "EVENT"], "gold": {"MUSIC_GENRE": ["Minneapolis sound", "funk", "R & B", "Rock music", "Pop music", "New wave music"], "BAND": ["the Time"], "MUSICAL_ARTIST": ["Prince"], "EVENT": [], "COUNTRY": [], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, BAND, MUSICAL_ARTIST, EVENT, COUNTRY, SONG\n\nText:\nPrince formed the Time, originally conceived as an opening act for him and based on his Minneapolis sound, a hybrid mixture of funk, R & B, Rock music, Pop music & amp; New wave music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, BAND, MUSICAL_ARTIST, EVENT, COUNTRY, SONG\n\nText:\nPrince formed the Time, originally conceived as an opening act for him and based on his Minneapolis sound, a hybrid mixture of funk, R & B, Rock music, Pop music & amp; New wave music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_38", "text": "Some managers in Europe soon created their own acts after being inspired by New Kids on the Block, beginning with Nigel Martin-Smith 's Take That in the UK (formed in 1990) and followed by Tom Watkins, who had success with Bros in the late 1980s and formed East 17 in 1991.", "requested_types": ["COUNTRY", "BAND", "ALBUM", "SONG", "LOCATION", "MUSICAL_ARTIST"], "distractors": ["SONG", "ALBUM"], "gold": {"COUNTRY": ["UK"], "BAND": ["New Kids on the Block", "Take That", "Bros", "East 17"], "ALBUM": [], "SONG": [], "LOCATION": ["Europe"], "MUSICAL_ARTIST": ["Nigel Martin-Smith", "Tom Watkins"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM, SONG, LOCATION, MUSICAL_ARTIST\n\nText:\nSome managers in Europe soon created their own acts after being inspired by New Kids on the Block, beginning with Nigel Martin-Smith 's Take That in the UK (formed in 1990) and followed by Tom Watkins, who had success with Bros in the late 1980s and formed East 17 in 1991.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM, SONG, LOCATION, MUSICAL_ARTIST\n\nText:\nSome managers in Europe soon created their own acts after being inspired by New Kids on the Block, beginning with Nigel Martin-Smith 's Take That in the UK (formed in 1990) and followed by Tom Watkins, who had success with Bros in the late 1980s and formed East 17 in 1991.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_39", "text": "It contained what would become Brooks' signature song, the blue collar anthem Friends in Low Places, as well as other popular singles, The Thunder Rolls and Unanswered Prayers.", "requested_types": ["LOCATION", "SONG", "MUSICAL_ARTIST"], "distractors": ["LOCATION"], "gold": {"LOCATION": [], "SONG": ["blue collar anthem", "Friends in Low Places", "The Thunder Rolls", "Unanswered Prayers"], "MUSICAL_ARTIST": ["Brooks"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST\n\nText:\nIt contained what would become Brooks' signature song, the blue collar anthem Friends in Low Places, as well as other popular singles, The Thunder Rolls and Unanswered Prayers.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST\n\nText:\nIt contained what would become Brooks' signature song, the blue collar anthem Friends in Low Places, as well as other popular singles, The Thunder Rolls and Unanswered Prayers.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_40", "text": "On 26 March 2001, their first full-length album, the self-titled Gorillaz, was released, producing four singles: Clint Eastwood, 19-2000, Rock the House, and Tomorrow Comes Today.", "requested_types": ["SONG", "BAND", "ALBUM"], "distractors": ["BAND"], "gold": {"SONG": ["Clint Eastwood", "19-2000", "Rock the House", "Tomorrow Comes Today"], "BAND": [], "ALBUM": ["Gorillaz"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, ALBUM\n\nText:\nOn 26 March 2001, their first full-length album, the self-titled Gorillaz, was released, producing four singles: Clint Eastwood, 19-2000, Rock the House, and Tomorrow Comes Today.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, ALBUM\n\nText:\nOn 26 March 2001, their first full-length album, the self-titled Gorillaz, was released, producing four singles: Clint Eastwood, 19-2000, Rock the House, and Tomorrow Comes Today.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_41", "text": "Such important centres include Mexico, the Andean region, Venezuela, and Paraguay.", "requested_types": ["LOCATION", "COUNTRY", "PERSON", "MUSICAL_INSTRUMENT", "MISC"], "distractors": ["MISC", "MUSICAL_INSTRUMENT", "PERSON"], "gold": {"LOCATION": ["Andean"], "COUNTRY": ["Mexico", "Venezuela", "Paraguay"], "PERSON": [], "MUSICAL_INSTRUMENT": [], "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: LOCATION, COUNTRY, PERSON, MUSICAL_INSTRUMENT, MISC\n\nText:\nSuch important centres include Mexico, the Andean region, Venezuela, and Paraguay.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, PERSON, MUSICAL_INSTRUMENT, MISC\n\nText:\nSuch important centres include Mexico, the Andean region, Venezuela, and Paraguay.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_42", "text": "By the end of the decade, Underwood had amassed eight No. 1 songs on the Billboard Hot Country Songs chart, along with numerous awards from the Country Music Association, Academy of Country Music and others.", "requested_types": ["MUSICAL_ARTIST", "AWARD", "EVENT", "ORGANISATION", "MISC", "SONG"], "distractors": ["AWARD", "EVENT", "SONG"], "gold": {"MUSICAL_ARTIST": ["Underwood"], "AWARD": [], "EVENT": [], "ORGANISATION": ["Billboard", "Country Music Association", "Academy of Country Music"], "MISC": ["Hot Country Songs chart"], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, AWARD, EVENT, ORGANISATION, MISC, SONG\n\nText:\nBy the end of the decade, Underwood had amassed eight No. 1 songs on the Billboard Hot Country Songs chart, along with numerous awards from the Country Music Association, Academy of Country Music 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: MUSICAL_ARTIST, AWARD, EVENT, ORGANISATION, MISC, SONG\n\nText:\nBy the end of the decade, Underwood had amassed eight No. 1 songs on the Billboard Hot Country Songs chart, along with numerous awards from the Country Music Association, Academy of Country Music 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_music_43", "text": "It was not until 1997 and the change to pop-oriented groups such as Backstreet Boys, 98 Degrees, NSYNC, The Moffatts, and Hanson that boy bands exploded commercially and dominated the market in the United States.", "requested_types": ["MUSIC_GENRE", "LOCATION", "SONG", "BAND", "COUNTRY"], "distractors": ["SONG", "LOCATION"], "gold": {"MUSIC_GENRE": ["pop-oriented"], "LOCATION": [], "SONG": [], "BAND": ["Backstreet Boys", "98 Degrees", "NSYNC", "The Moffatts", "Hanson"], "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: MUSIC_GENRE, LOCATION, SONG, BAND, COUNTRY\n\nText:\nIt was not until 1997 and the change to pop-oriented groups such as Backstreet Boys, 98 Degrees, NSYNC, The Moffatts, and Hanson that boy bands exploded commercially and dominated the market 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: MUSIC_GENRE, LOCATION, SONG, BAND, COUNTRY\n\nText:\nIt was not until 1997 and the change to pop-oriented groups such as Backstreet Boys, 98 Degrees, NSYNC, The Moffatts, and Hanson that boy bands exploded commercially and dominated the market 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_music_44", "text": "The National Squash Centre and the National Cycling Centre, which includes both the Manchester Velodrome and the National Indoor BMX Arena, are all a short distance from the stadium.", "requested_types": ["MUSICAL_INSTRUMENT", "LOCATION", "COUNTRY"], "distractors": ["MUSICAL_INSTRUMENT", "COUNTRY"], "gold": {"MUSICAL_INSTRUMENT": [], "LOCATION": ["National Squash Centre", "National Cycling Centre", "Manchester Velodrome", "National Indoor BMX Arena"], "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: MUSICAL_INSTRUMENT, LOCATION, COUNTRY\n\nText:\nThe National Squash Centre and the National Cycling Centre, which includes both the Manchester Velodrome and the National Indoor BMX Arena, are all a short distance from the stadium.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, LOCATION, COUNTRY\n\nText:\nThe National Squash Centre and the National Cycling Centre, which includes both the Manchester Velodrome and the National Indoor BMX Arena, are all a short distance from the stadium.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_45", "text": "The film was a success, securing three Oscar nominations, including Academy Award for Best Picture, Academy Award for Best Sound Mixing and Academy Award for Best Supporting Actor for Lemmon, who won.", "requested_types": ["PERSON", "MUSIC_GENRE", "AWARD"], "distractors": ["MUSIC_GENRE"], "gold": {"PERSON": ["Lemmon"], "MUSIC_GENRE": [], "AWARD": ["Oscar", "Academy Award for Best Picture", "Academy Award for Best Sound Mixing", "Academy Award for Best Supporting Actor"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, AWARD\n\nText:\nThe film was a success, securing three Oscar nominations, including Academy Award for Best Picture, Academy Award for Best Sound Mixing and Academy Award for Best Supporting Actor for Lemmon, who won.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, AWARD\n\nText:\nThe film was a success, securing three Oscar nominations, including Academy Award for Best Picture, Academy Award for Best Sound Mixing and Academy Award for Best Supporting Actor for Lemmon, who won.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_46", "text": "In January 1970 Lucian K. Truscott IV reviewing Led Zeppelin II for the Village Voice described the sound as heavy and made comparisons with Blue Cheer and Vanilla Fudge.. January 22, 1970.", "requested_types": ["AWARD", "MISC", "BAND", "PERSON", "ALBUM"], "distractors": ["AWARD"], "gold": {"AWARD": [], "MISC": ["Village Voice"], "BAND": ["Blue Cheer", "Vanilla Fudge"], "PERSON": ["Lucian K. Truscott IV"], "ALBUM": ["Led Zeppelin 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: AWARD, MISC, BAND, PERSON, ALBUM\n\nText:\nIn January 1970 Lucian K. Truscott IV reviewing Led Zeppelin II for the Village Voice described the sound as heavy and made comparisons with Blue Cheer and Vanilla Fudge.. January 22, 1970.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, PERSON, ALBUM\n\nText:\nIn January 1970 Lucian K. Truscott IV reviewing Led Zeppelin II for the Village Voice described the sound as heavy and made comparisons with Blue Cheer and Vanilla Fudge.. January 22, 1970.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_47", "text": "The success and popularity of the Commonwealth Games resulted in Edmonton bidding for and being selected to host the 1983 Summer Universiade.", "requested_types": ["SONG", "LOCATION", "AWARD", "EVENT"], "distractors": ["AWARD", "SONG"], "gold": {"SONG": [], "LOCATION": ["Edmonton"], "AWARD": [], "EVENT": ["Commonwealth Games", "1983 Summer Universiade"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, LOCATION, AWARD, EVENT\n\nText:\nThe success and popularity of the Commonwealth Games resulted in Edmonton bidding for and being selected to host the 1983 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: SONG, LOCATION, AWARD, EVENT\n\nText:\nThe success and popularity of the Commonwealth Games resulted in Edmonton bidding for and being selected to host the 1983 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_music_48", "text": "In addition to being named Top Female Vocalist by the Academy of Country Music (ACM) twice and Female Vocalist of the Year by the Country Music Association (CMA), she also won a Grammy Award (earning seven nominations), People's Choice Award and an American Music Award (AMA).", "requested_types": ["AWARD", "MISC", "COUNTRY", "ORGANISATION", "PERSON"], "distractors": ["MISC", "PERSON", "COUNTRY"], "gold": {"AWARD": ["Top Female Vocalist", "Female Vocalist of the Year", "Grammy Award", "People's Choice Award", "American Music Award", "AMA"], "MISC": [], "COUNTRY": [], "ORGANISATION": ["Academy of Country Music", "ACM", "Country Music Association", "CMA"], "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: AWARD, MISC, COUNTRY, ORGANISATION, PERSON\n\nText:\nIn addition to being named Top Female Vocalist by the Academy of Country Music (ACM) twice and Female Vocalist of the Year by the Country Music Association (CMA), she also won a Grammy Award (earning seven nominations), People's Choice Award and an American Music Award (AMA).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, ORGANISATION, PERSON\n\nText:\nIn addition to being named Top Female Vocalist by the Academy of Country Music (ACM) twice and Female Vocalist of the Year by the Country Music Association (CMA), she also won a Grammy Award (earning seven nominations), People's Choice Award and an American Music Award (AMA).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_49", "text": "In 1985, the trio joined up with Marc Almond to record a version of Donna Summer 's I Feel Love.", "requested_types": ["SONG", "MUSICAL_ARTIST", "EVENT", "PERSON", "BAND"], "distractors": ["EVENT", "PERSON"], "gold": {"SONG": ["I Feel Love"], "MUSICAL_ARTIST": ["Marc Almond", "Donna Summer"], "EVENT": [], "PERSON": [], "BAND": ["trio"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, EVENT, PERSON, BAND\n\nText:\nIn 1985, the trio joined up with Marc Almond to record a version of Donna Summer 's I Feel Love.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, EVENT, PERSON, BAND\n\nText:\nIn 1985, the trio joined up with Marc Almond to record a version of Donna Summer 's I Feel Love.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_50", "text": "They won a Grammy Award in 1999 in the Grammy Award for Best Polka Album category for their album Polkasonic, and again in 2004 for their album Let's Kiss.", "requested_types": ["ALBUM", "MISC", "LOCATION", "BAND", "AWARD"], "distractors": ["MISC", "LOCATION", "BAND"], "gold": {"ALBUM": ["Polkasonic", "Let's Kiss"], "MISC": [], "LOCATION": [], "BAND": [], "AWARD": ["Grammy Award", "Grammy Award for Best Polka Album"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MISC, LOCATION, BAND, AWARD\n\nText:\nThey won a Grammy Award in 1999 in the Grammy Award for Best Polka Album category for their album Polkasonic, and again in 2004 for their album Let's Kiss.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MISC, LOCATION, BAND, AWARD\n\nText:\nThey won a Grammy Award in 1999 in the Grammy Award for Best Polka Album category for their album Polkasonic, and again in 2004 for their album Let's Kiss.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_51", "text": "Harper also contributed a recital of Jabberwocky for The Wildlife Album, an 18-track compilation CD to benefit the World Wide Fund for Nature and the Ulster Wildlife Trust.", "requested_types": ["MUSICAL_ARTIST", "SONG", "MISC", "ALBUM", "ORGANISATION", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"MUSICAL_ARTIST": ["Harper"], "SONG": ["Jabberwocky"], "MISC": ["CD"], "ALBUM": ["Wildlife Album"], "ORGANISATION": ["World Wide Fund for Nature", "Ulster Wildlife Trust"], "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: MUSICAL_ARTIST, SONG, MISC, ALBUM, ORGANISATION, COUNTRY\n\nText:\nHarper also contributed a recital of Jabberwocky for The Wildlife Album, an 18-track compilation CD to benefit the World Wide Fund for Nature and the Ulster Wildlife Trust.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, SONG, MISC, ALBUM, ORGANISATION, COUNTRY\n\nText:\nHarper also contributed a recital of Jabberwocky for The Wildlife Album, an 18-track compilation CD to benefit the World Wide Fund for Nature and the Ulster Wildlife Trust.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_52", "text": "After their first two albums, Fungus Amongus (1995) and S.C.I.E.N.C.E. (1997), the band earned mainstream recognition with the release of their 1999 album Make Yourself which spawned several hits, including the band's highest charting song Drive.", "requested_types": ["MISC", "ALBUM", "SONG"], "distractors": ["MISC"], "gold": {"MISC": [], "ALBUM": ["Fungus Amongus", "S.C.I.E.N.C.E.", "Make Yourself"], "SONG": ["Drive"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, SONG\n\nText:\nAfter their first two albums, Fungus Amongus (1995) and S.C.I.E.N.C.E. (1997), the band earned mainstream recognition with the release of their 1999 album Make Yourself which spawned several hits, including the band's highest charting song Drive.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, SONG\n\nText:\nAfter their first two albums, Fungus Amongus (1995) and S.C.I.E.N.C.E. (1997), the band earned mainstream recognition with the release of their 1999 album Make Yourself which spawned several hits, including the band's highest charting song Drive.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_53", "text": "Funk drumming creates a groove by emphasizing the drummer's feel and emotion, which including occasional tempo fluctuations, the use of Swing music feel in some songs (e.g., Cissy Strut by The Meters and I'll Take You There by The Staple Singers, which have a half-swung feel), and less use of fills (as they can lessen the groove).", "requested_types": ["SONG", "ORGANISATION", "MISC", "MUSIC_GENRE", "BAND"], "distractors": ["ORGANISATION"], "gold": {"SONG": ["Cissy Strut", "I'll Take You There"], "ORGANISATION": [], "MISC": ["Funk drumming"], "MUSIC_GENRE": ["Swing music"], "BAND": ["The Meters", "The Staple Singers"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, ORGANISATION, MISC, MUSIC_GENRE, BAND\n\nText:\nFunk drumming creates a groove by emphasizing the drummer's feel and emotion, which including occasional tempo fluctuations, the use of Swing music feel in some songs (e.g., Cissy Strut by The Meters and I'll Take You There by The Staple Singers, which have a half-swung feel), and less use of fills (as they can lessen the groove).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, ORGANISATION, MISC, MUSIC_GENRE, BAND\n\nText:\nFunk drumming creates a groove by emphasizing the drummer's feel and emotion, which including occasional tempo fluctuations, the use of Swing music feel in some songs (e.g., Cissy Strut by The Meters and I'll Take You There by The Staple Singers, which have a half-swung feel), and less use of fills (as they can lessen the groove).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_54", "text": "He has earned numerous awards for his work, including 20 Grammy Award s, two Golden Globe s, an Academy Awards, and a Tony Award (for Springsteen on Broadway).", "requested_types": ["AWARD", "SONG", "ALBUM", "MISC", "COUNTRY"], "distractors": ["MISC", "COUNTRY", "SONG"], "gold": {"AWARD": ["Grammy Award", "Golden Globe", "Academy Awards", "Tony Award"], "SONG": [], "ALBUM": ["Springsteen on Broadway"], "MISC": [], "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, SONG, ALBUM, MISC, COUNTRY\n\nText:\nHe has earned numerous awards for his work, including 20 Grammy Award s, two Golden Globe s, an Academy Awards, and a Tony Award (for Springsteen on Broadway).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, ALBUM, MISC, COUNTRY\n\nText:\nHe has earned numerous awards for his work, including 20 Grammy Award s, two Golden Globe s, an Academy Awards, and a Tony Award (for Springsteen on Broadway).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_55", "text": "The album also won Best Album at the 2002 MTV Europe Music Awards, Billboard Music Award for Top Billboard 200 Album at the 2002 Billboard Music Awards and Best International Album and International Album of the Year at the Brit Awards and the Juno Awards respectively in 2003.", "requested_types": ["AWARD", "EVENT", "MUSICAL_INSTRUMENT"], "distractors": ["MUSICAL_INSTRUMENT", "EVENT"], "gold": {"AWARD": ["Best Album at the 2002 MTV Europe Music Awards", "Billboard Music Award for Top Billboard 200 Album", "Billboard Music Awards", "Best International Album", "International Album of the Year", "Brit Awards", "Juno Awards"], "EVENT": [], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT, MUSICAL_INSTRUMENT\n\nText:\nThe album also won Best Album at the 2002 MTV Europe Music Awards, Billboard Music Award for Top Billboard 200 Album at the 2002 Billboard Music Awards and Best International Album and International Album of the Year at the Brit Awards and the Juno Awards respectively in 2003.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT, MUSICAL_INSTRUMENT\n\nText:\nThe album also won Best Album at the 2002 MTV Europe Music Awards, Billboard Music Award for Top Billboard 200 Album at the 2002 Billboard Music Awards and Best International Album and International Album of the Year at the Brit Awards and the Juno Awards respectively in 2003.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_56", "text": "During this period, Freddie King and Albert King often played with rock and Soul music musicians (Eric Clapton and Booker T & the MGs) and had a major influence on those styles of music.", "requested_types": ["SONG", "BAND", "MUSICAL_ARTIST", "MUSIC_GENRE", "COUNTRY"], "distractors": ["SONG", "COUNTRY"], "gold": {"SONG": [], "BAND": ["Booker T & the MGs"], "MUSICAL_ARTIST": ["Freddie King", "Albert King", "Eric Clapton"], "MUSIC_GENRE": ["rock", "Soul music"], "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: SONG, BAND, MUSICAL_ARTIST, MUSIC_GENRE, COUNTRY\n\nText:\nDuring this period, Freddie King and Albert King often played with rock and Soul music musicians (Eric Clapton and Booker T & the MGs) and had a major influence on those styles of music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, MUSICAL_ARTIST, MUSIC_GENRE, COUNTRY\n\nText:\nDuring this period, Freddie King and Albert King often played with rock and Soul music musicians (Eric Clapton and Booker T & the MGs) and had a major influence on those styles of music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_57", "text": "1998 brought a sequel, Blues Brothers 2000 that, while not holding as great a critical and financial success, featured a much larger number of blues artists, such as B.B. King, Bo Diddley, Erykah Badu, Eric Clapton, Steve Winwood, Charlie Musselwhite, Blues Traveler, Jimmie Vaughan, and Jeff Baxter.", "requested_types": ["BAND", "MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "EVENT", "PERSON", "MISC", "MUSIC_GENRE"], "distractors": ["EVENT", "MUSICAL_INSTRUMENT", "PERSON"], "gold": {"BAND": ["Blues Traveler"], "MUSICAL_ARTIST": ["B.B. King", "Bo Diddley", "Erykah Badu", "Eric Clapton", "Steve Winwood", "Charlie Musselwhite", "Jimmie Vaughan", "Jeff Baxter"], "MUSICAL_INSTRUMENT": [], "EVENT": [], "PERSON": [], "MISC": ["Blues Brothers 2000"], "MUSIC_GENRE": ["blues"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, EVENT, PERSON, MISC, MUSIC_GENRE\n\nText:\n1998 brought a sequel, Blues Brothers 2000 that, while not holding as great a critical and financial success, featured a much larger number of blues artists, such as B.B. King, Bo Diddley, Erykah Badu, Eric Clapton, Steve Winwood, Charlie Musselwhite, Blues Traveler, Jimmie Vaughan, and Jeff Baxter.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, EVENT, PERSON, MISC, MUSIC_GENRE\n\nText:\n1998 brought a sequel, Blues Brothers 2000 that, while not holding as great a critical and financial success, featured a much larger number of blues artists, such as B.B. King, Bo Diddley, Erykah Badu, Eric Clapton, Steve Winwood, Charlie Musselwhite, Blues Traveler, Jimmie Vaughan, and Jeff Baxter.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_58", "text": "It includes collaborations with Pete Seeger, Ivan Neville, Cyril Neville, Skerik, Adam Levy, Righteous Babe recording artist Anaïs Mitchell, CC Adcock, and a host of New Orleans-based horn players known for their work in such outfits as Galactic, Bonerama, and Rebirth Brass Band.", "requested_types": ["MISC", "MUSICAL_ARTIST", "EVENT", "BAND", "ORGANISATION"], "distractors": ["EVENT"], "gold": {"MISC": ["New", "Orleans-based"], "MUSICAL_ARTIST": ["Pete Seeger", "Ivan Neville", "Cyril Neville", "Skerik", "Adam Levy", "Anaïs Mitchell", "CC Adcock"], "EVENT": [], "BAND": ["Galactic", "Bonerama", "Rebirth Brass Band"], "ORGANISATION": ["Righteous Babe recording"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, EVENT, BAND, ORGANISATION\n\nText:\nIt includes collaborations with Pete Seeger, Ivan Neville, Cyril Neville, Skerik, Adam Levy, Righteous Babe recording artist Anaïs Mitchell, CC Adcock, and a host of New Orleans-based horn players known for their work in such outfits as Galactic, Bonerama, and Rebirth Brass Band.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, EVENT, BAND, ORGANISATION\n\nText:\nIt includes collaborations with Pete Seeger, Ivan Neville, Cyril Neville, Skerik, Adam Levy, Righteous Babe recording artist Anaïs Mitchell, CC Adcock, and a host of New Orleans-based horn players known for their work in such outfits as Galactic, Bonerama, and Rebirth Brass Band.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_59", "text": "Indo-Caribbean music of Indo-Caribbean people in Caribbean is most common in Trinidad and Tobago, Guyana, Jamaica, and Suriname, which reflects their Bhojpuri heritage.", "requested_types": ["MISC", "LOCATION", "BAND", "COUNTRY", "SONG", "ALBUM"], "distractors": ["BAND", "ALBUM", "SONG"], "gold": {"MISC": ["Indo-Caribbean", "Bhojpuri"], "LOCATION": ["Caribbean"], "BAND": [], "COUNTRY": ["Trinidad and Tobago", "Guyana", "Jamaica", "Suriname"], "SONG": [], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, COUNTRY, SONG, ALBUM\n\nText:\nIndo-Caribbean music of Indo-Caribbean people in Caribbean is most common in Trinidad and Tobago, Guyana, Jamaica, and Suriname, which reflects their Bhojpuri heritage.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, COUNTRY, SONG, ALBUM\n\nText:\nIndo-Caribbean music of Indo-Caribbean people in Caribbean is most common in Trinidad and Tobago, Guyana, Jamaica, and Suriname, which reflects their Bhojpuri heritage.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_60", "text": "The New Orleans setting of the film played to Newman's musical strengths, and his songs contained elements of Cajun music, zydeco, blues and Dixieland.", "requested_types": ["AWARD", "PERSON", "MUSICAL_ARTIST", "EVENT", "MUSIC_GENRE"], "distractors": ["EVENT", "AWARD", "PERSON"], "gold": {"AWARD": [], "PERSON": [], "MUSICAL_ARTIST": ["Newman"], "EVENT": [], "MUSIC_GENRE": ["New Orleans", "Cajun music", "zydeco", "blues", "Dixieland"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, EVENT, MUSIC_GENRE\n\nText:\nThe New Orleans setting of the film played to Newman's musical strengths, and his songs contained elements of Cajun music, zydeco, blues and Dixieland.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, EVENT, MUSIC_GENRE\n\nText:\nThe New Orleans setting of the film played to Newman's musical strengths, and his songs contained elements of Cajun music, zydeco, blues and Dixieland.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_61", "text": "They also released five singles to promote the album: Enter Sandman, The Unforgiven, Nothing Else Matters, Wherever I May Roam, and Sad but TRUE, all of which have been considered to be among the band's best-known songs.", "requested_types": ["SONG", "MISC"], "distractors": ["MISC"], "gold": {"SONG": ["Enter Sandman", "The Unforgiven", "Nothing Else Matters", "Wherever I May Roam", "Sad but TRUE"], "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: SONG, MISC\n\nText:\nThey also released five singles to promote the album: Enter Sandman, The Unforgiven, Nothing Else Matters, Wherever I May Roam, and Sad but TRUE, all of which have been considered to be among the band's best-known songs.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MISC\n\nText:\nThey also released five singles to promote the album: Enter Sandman, The Unforgiven, Nothing Else Matters, Wherever I May Roam, and Sad but TRUE, all of which have been considered to be among the band's best-known songs.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_62", "text": "He helped found the American Conservatory Theater in San Francisco, the Mark Taper Forum in Los Angeles, and the Brooklyn Academy of Music Repertory Company in New York City.", "requested_types": ["LOCATION", "ALBUM"], "distractors": ["ALBUM"], "gold": {"LOCATION": ["American Conservatory Theater", "San Francisco", "Mark Taper Forum", "Los Angeles", "Brooklyn Academy of Music", "Repertory Company", "New York City"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nHe helped found the American Conservatory Theater in San Francisco, the Mark Taper Forum in Los Angeles, and the Brooklyn Academy of Music Repertory Company in New York 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, ALBUM\n\nText:\nHe helped found the American Conservatory Theater in San Francisco, the Mark Taper Forum in Los Angeles, and the Brooklyn Academy of Music Repertory Company in New York 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_music_63", "text": "Cliff Bruner, Moon Mullican, Milton Brown and Adolph Hofner were other early Western swing pioneers.", "requested_types": ["MISC", "SONG", "AWARD", "MUSIC_GENRE", "MUSICAL_ARTIST"], "distractors": ["MISC", "AWARD", "SONG"], "gold": {"MISC": [], "SONG": [], "AWARD": [], "MUSIC_GENRE": ["Western swing"], "MUSICAL_ARTIST": ["Cliff Bruner", "Moon Mullican", "Milton Brown", "Adolph Hofner"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, AWARD, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nCliff Bruner, Moon Mullican, Milton Brown and Adolph Hofner were other early Western swing pioneers.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, AWARD, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nCliff Bruner, Moon Mullican, Milton Brown and Adolph Hofner were other early Western swing pioneers.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_64", "text": "The Police won a number of music awards, including six Grammy Award s, two Brit Awards - winning Best British Group once, an MTV Video Music Award, and in 2003 were inducted into the Rock and Roll Hall of Fame.. Rolling Stone.", "requested_types": ["AWARD", "BAND", "ORGANISATION", "MISC"], "distractors": ["MISC"], "gold": {"AWARD": ["Grammy Award", "Brit Awards", "Best British Group", "MTV Video Music Award"], "BAND": ["The Police"], "ORGANISATION": ["Rock and Roll Hall of Fame", "Rolling Stone"], "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, BAND, ORGANISATION, MISC\n\nText:\nThe Police won a number of music awards, including six Grammy Award s, two Brit Awards - winning Best British Group once, an MTV Video Music Award, and in 2003 were inducted into the Rock and Roll Hall of Fame.. Rolling Stone.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ORGANISATION, MISC\n\nText:\nThe Police won a number of music awards, including six Grammy Award s, two Brit Awards - winning Best British Group once, an MTV Video Music Award, and in 2003 were inducted into the Rock and Roll Hall of Fame.. Rolling Stone.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_65", "text": "Alt-country, in various iterations overlapped with other genres, including Red Dirt country music (Cross Canadian Ragweed), jam band s (My Morning Jacket and The String Cheese Incident), and indie folk (The Avett Brothers).", "requested_types": ["MUSICAL_INSTRUMENT", "MUSIC_GENRE", "MUSICAL_ARTIST", "BAND", "MISC"], "distractors": ["MUSICAL_INSTRUMENT", "MUSICAL_ARTIST"], "gold": {"MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": ["Alt-country", "Red Dirt country music", "indie folk"], "MUSICAL_ARTIST": [], "BAND": ["Cross Canadian Ragweed", "My Morning Jacket", "The String Cheese Incident", "The Avett Brothers"], "MISC": ["jam band"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSIC_GENRE, MUSICAL_ARTIST, BAND, MISC\n\nText:\nAlt-country, in various iterations overlapped with other genres, including Red Dirt country music (Cross Canadian Ragweed), jam band s (My Morning Jacket and The String Cheese Incident), and indie folk (The Avett Brothers).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSIC_GENRE, MUSICAL_ARTIST, BAND, MISC\n\nText:\nAlt-country, in various iterations overlapped with other genres, including Red Dirt country music (Cross Canadian Ragweed), jam band s (My Morning Jacket and The String Cheese Incident), and indie folk (The Avett Brothers).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_66", "text": "Examples of this include Michael Jackson from The Jackson 5, Donny Osmond from The Osmonds, Ricky Martin from Menudo, Justin Timberlake from * NSYNC, and Ronan Keating from Boyzone.", "requested_types": ["MUSICAL_ARTIST", "MISC", "ORGANISATION", "PERSON", "BAND", "COUNTRY"], "distractors": ["MISC", "COUNTRY", "ORGANISATION"], "gold": {"MUSICAL_ARTIST": ["Michael Jackson", "Donny Osmond"], "MISC": [], "ORGANISATION": [], "PERSON": ["Ricky Martin", "Justin Timberlake", "Ronan Keating"], "BAND": ["The Jackson 5", "The Osmonds", "Menudo", "NSYNC", "Boyzone"], "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: MUSICAL_ARTIST, MISC, ORGANISATION, PERSON, BAND, COUNTRY\n\nText:\nExamples of this include Michael Jackson from The Jackson 5, Donny Osmond from The Osmonds, Ricky Martin from Menudo, Justin Timberlake from * NSYNC, and Ronan Keating from Boyzone.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MISC, ORGANISATION, PERSON, BAND, COUNTRY\n\nText:\nExamples of this include Michael Jackson from The Jackson 5, Donny Osmond from The Osmonds, Ricky Martin from Menudo, Justin Timberlake from * NSYNC, and Ronan Keating from Boyzone.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_67", "text": "Baduizm was certified three times platinum by the Recording Industry Association of America, Gold by the British Phonographic Industry and the Canadian Recording Industry Association.", "requested_types": ["ALBUM", "ORGANISATION", "LOCATION", "BAND"], "distractors": ["BAND", "LOCATION"], "gold": {"ALBUM": ["Baduizm"], "ORGANISATION": ["Recording Industry Association of America", "British Phonographic Industry", "Canadian Recording Industry Association"], "LOCATION": [], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, ORGANISATION, LOCATION, BAND\n\nText:\nBaduizm was certified three times platinum by the Recording Industry Association of America, Gold by the British Phonographic Industry and the Canadian Recording Industry 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: ALBUM, ORGANISATION, LOCATION, BAND\n\nText:\nBaduizm was certified three times platinum by the Recording Industry Association of America, Gold by the British Phonographic Industry and the Canadian Recording Industry 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_music_68", "text": "Dead Kennedys have influenced multiple acts such as System of a Down, Green Day, Faith No More, Rage Against the Machine, Anthrax, Sepultura, Descendents, Bad Religion, X, Minutemen, The Hives, Saves the Day and Screeching Weasel among others.", "requested_types": ["BAND", "SONG", "COUNTRY", "ALBUM"], "distractors": ["COUNTRY", "SONG", "ALBUM"], "gold": {"BAND": ["Dead Kennedys", "System of a Down", "Green Day", "Faith No More", "Rage Against the Machine", "Anthrax", "Sepultura", "Descendents", "Bad Religion", "X", "Minutemen", "The Hives", "Saves the Day", "Screeching Weasel"], "SONG": [], "COUNTRY": [], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, SONG, COUNTRY, ALBUM\n\nText:\nDead Kennedys have influenced multiple acts such as System of a Down, Green Day, Faith No More, Rage Against the Machine, Anthrax, Sepultura, Descendents, Bad Religion, X, Minutemen, The Hives, Saves the Day and Screeching Weasel 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: BAND, SONG, COUNTRY, ALBUM\n\nText:\nDead Kennedys have influenced multiple acts such as System of a Down, Green Day, Faith No More, Rage Against the Machine, Anthrax, Sepultura, Descendents, Bad Religion, X, Minutemen, The Hives, Saves the Day and Screeching Weasel 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_music_69", "text": "Today, zydeco integrates genres such as Rhythm and blues, Soul music, brass band, reggae, Hip hop music, ska, Rock music, Afro-Caribbean music and other styles, in addition to the traditional forms.", "requested_types": ["MUSIC_GENRE", "ORGANISATION"], "distractors": ["ORGANISATION"], "gold": {"MUSIC_GENRE": ["Rhythm and blues", "Soul music", "brass band", "reggae", "Hip hop music", "ska", "Rock music", "Afro-Caribbean music"], "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: MUSIC_GENRE, ORGANISATION\n\nText:\nToday, zydeco integrates genres such as Rhythm and blues, Soul music, brass band, reggae, Hip hop music, ska, Rock music, Afro-Caribbean music and other styles, in addition to the traditional forms.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, ORGANISATION\n\nText:\nToday, zydeco integrates genres such as Rhythm and blues, Soul music, brass band, reggae, Hip hop music, ska, Rock music, Afro-Caribbean music and other styles, in addition to the traditional forms.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_70", "text": "The stadium will be the opening and closing ceremonies, athletics, and football venue for the 2014 South American Games and the 2023 Pan American Games.", "requested_types": ["MUSIC_GENRE", "SONG", "EVENT"], "distractors": ["SONG", "MUSIC_GENRE"], "gold": {"MUSIC_GENRE": [], "SONG": [], "EVENT": ["2014 South American Games", "2023 Pan American 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: MUSIC_GENRE, SONG, EVENT\n\nText:\nThe stadium will be the opening and closing ceremonies, athletics, and football venue for the 2014 South American Games and the 2023 Pan American 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: MUSIC_GENRE, SONG, EVENT\n\nText:\nThe stadium will be the opening and closing ceremonies, athletics, and football venue for the 2014 South American Games and the 2023 Pan American 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_music_71", "text": "There have also been several spin-offs featuring one or more core members, such as Dead & amp; Company, Furthur, the Rhythm Devils, Phil Lesh and Friends, RatDog, and Billy & amp; the Kids.", "requested_types": ["LOCATION", "MUSICAL_ARTIST", "BAND", "COUNTRY"], "distractors": ["MUSICAL_ARTIST", "LOCATION", "COUNTRY"], "gold": {"LOCATION": [], "MUSICAL_ARTIST": [], "BAND": ["Dead & amp; Company", "Furthur", "Rhythm Devils", "Phil Lesh and Friends", "RatDog", "Billy & amp; the Kids"], "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: LOCATION, MUSICAL_ARTIST, BAND, COUNTRY\n\nText:\nThere have also been several spin-offs featuring one or more core members, such as Dead & amp; Company, Furthur, the Rhythm Devils, Phil Lesh and Friends, RatDog, and Billy & amp; the Kids.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, BAND, COUNTRY\n\nText:\nThere have also been several spin-offs featuring one or more core members, such as Dead & amp; Company, Furthur, the Rhythm Devils, Phil Lesh and Friends, RatDog, and Billy & amp; the Kids.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_72", "text": "White audiences' interest in the blues during the 1960s increased due to the Chicago-based Paul Butterfield featuring guitarist Mike Bloomfield, and the British blues movement.", "requested_types": ["PERSON", "ALBUM", "MUSIC_GENRE", "LOCATION", "EVENT", "MUSICAL_ARTIST"], "distractors": ["PERSON", "ALBUM"], "gold": {"PERSON": [], "ALBUM": [], "MUSIC_GENRE": ["blues"], "LOCATION": ["Chicago-based"], "EVENT": ["British blues movement"], "MUSICAL_ARTIST": ["Paul Butterfield", "Mike Bloomfield"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MUSIC_GENRE, LOCATION, EVENT, MUSICAL_ARTIST\n\nText:\nWhite audiences' interest in the blues during the 1960s increased due to the Chicago-based Paul Butterfield featuring guitarist Mike Bloomfield, and the British blues movement.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MUSIC_GENRE, LOCATION, EVENT, MUSICAL_ARTIST\n\nText:\nWhite audiences' interest in the blues during the 1960s increased due to the Chicago-based Paul Butterfield featuring guitarist Mike Bloomfield, and the British blues movement.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_73", "text": "Little Walter, Sonny Boy Williamson II and Sonny Terry are well known harmonica (called harp by blues musicians) players of the early Chicago blues scene.", "requested_types": ["MUSICAL_ARTIST", "MUSIC_GENRE", "ALBUM", "MUSICAL_INSTRUMENT", "EVENT", "ORGANISATION"], "distractors": ["ORGANISATION", "ALBUM", "EVENT"], "gold": {"MUSICAL_ARTIST": ["Little Walter", "Sonny Boy Williamson II", "Sonny Terry"], "MUSIC_GENRE": ["blues", "Chicago blues"], "ALBUM": [], "MUSICAL_INSTRUMENT": ["harmonica", "harp"], "EVENT": [], "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: MUSICAL_ARTIST, MUSIC_GENRE, ALBUM, MUSICAL_INSTRUMENT, EVENT, ORGANISATION\n\nText:\nLittle Walter, Sonny Boy Williamson II and Sonny Terry are well known harmonica (called harp by blues musicians) players of the early Chicago blues scene.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MUSIC_GENRE, ALBUM, MUSICAL_INSTRUMENT, EVENT, ORGANISATION\n\nText:\nLittle Walter, Sonny Boy Williamson II and Sonny Terry are well known harmonica (called harp by blues musicians) players of the early Chicago blues scene.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_74", "text": "The territory band s operating out of Kansas City, the Bennie Moten orchestra, Jay McShann, and the Count Basie Orchestra were also concentrating on the blues, with 12-bar blues instrumentals such as Basie's One O'Clock Jump and Jumpin 'at the Woodside and boisterous blues shouting by Jimmy Rushing on songs such as Going to Chicago and Sent for You Yesterday.", "requested_types": ["BAND", "SONG", "ALBUM", "MUSIC_GENRE", "MISC", "PERSON", "LOCATION", "COUNTRY", "MUSICAL_ARTIST"], "distractors": ["PERSON", "ALBUM", "COUNTRY"], "gold": {"BAND": ["Bennie Moten orchestra", "Count Basie Orchestra"], "SONG": ["One O'Clock Jump", "Jumpin 'at the Woodside", "Going to Chicago", "Sent for You Yesterday"], "ALBUM": [], "MUSIC_GENRE": ["blues"], "MISC": ["territory band", "12-bar blues instrumentals"], "PERSON": [], "LOCATION": ["Kansas City"], "COUNTRY": [], "MUSICAL_ARTIST": ["Jay McShann", "Basie", "Jimmy Rushing"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, SONG, ALBUM, MUSIC_GENRE, MISC, PERSON, LOCATION, COUNTRY, MUSICAL_ARTIST\n\nText:\nThe territory band s operating out of Kansas City, the Bennie Moten orchestra, Jay McShann, and the Count Basie Orchestra were also concentrating on the blues, with 12-bar blues instrumentals such as Basie's One O'Clock Jump and Jumpin 'at the Woodside and boisterous blues shouting by Jimmy Rushing on songs such as Going to Chicago and Sent for You Yesterday.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, SONG, ALBUM, MUSIC_GENRE, MISC, PERSON, LOCATION, COUNTRY, MUSICAL_ARTIST\n\nText:\nThe territory band s operating out of Kansas City, the Bennie Moten orchestra, Jay McShann, and the Count Basie Orchestra were also concentrating on the blues, with 12-bar blues instrumentals such as Basie's One O'Clock Jump and Jumpin 'at the Woodside and boisterous blues shouting by Jimmy Rushing on songs such as Going to Chicago and Sent for You Yesterday.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_75", "text": "Troy Cassar-Daley is among Australia's successful contemporary indigenous performers, and Kev Carmody and Archie Roach employ a combination of folk-rock and country music to sing about Aboriginal rights issues.", "requested_types": ["MUSICAL_ARTIST", "ALBUM", "MISC", "SONG", "COUNTRY", "MUSIC_GENRE"], "distractors": ["ALBUM", "SONG"], "gold": {"MUSICAL_ARTIST": ["Troy Cassar-Daley", "Kev Carmody", "Archie Roach"], "ALBUM": [], "MISC": ["contemporary indigenous performers", "Aboriginal"], "SONG": [], "COUNTRY": ["Australia"], "MUSIC_GENRE": ["folk-rock", "country music"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ALBUM, MISC, SONG, COUNTRY, MUSIC_GENRE\n\nText:\nTroy Cassar-Daley is among Australia's successful contemporary indigenous performers, and Kev Carmody and Archie Roach employ a combination of folk-rock and country music to sing about Aboriginal rights issues.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ALBUM, MISC, SONG, COUNTRY, MUSIC_GENRE\n\nText:\nTroy Cassar-Daley is among Australia's successful contemporary indigenous performers, and Kev Carmody and Archie Roach employ a combination of folk-rock and country music to sing about Aboriginal rights issues.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_76", "text": "She won Top New Female Vocalist from the Academy of Country Music and the Horizon Award from the Country Music Association.", "requested_types": ["COUNTRY", "AWARD", "MUSICAL_ARTIST", "BAND", "ORGANISATION"], "distractors": ["COUNTRY", "MUSICAL_ARTIST", "BAND"], "gold": {"COUNTRY": [], "AWARD": ["Top New Female Vocalist", "Horizon Award"], "MUSICAL_ARTIST": [], "BAND": [], "ORGANISATION": ["Academy of Country Music", "Country Music 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: COUNTRY, AWARD, MUSICAL_ARTIST, BAND, ORGANISATION\n\nText:\nShe won Top New Female Vocalist from the Academy of Country Music and the Horizon Award from the Country Music 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: COUNTRY, AWARD, MUSICAL_ARTIST, BAND, ORGANISATION\n\nText:\nShe won Top New Female Vocalist from the Academy of Country Music and the Horizon Award from the Country Music 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_music_77", "text": "The band headlined the Soul Assassins tour with House of Pain and Funkdoobiest as support, then performed on a college tour with Rage Against the Machine and Seven Year Bitch.", "requested_types": ["BAND", "MISC", "ALBUM"], "distractors": ["ALBUM"], "gold": {"BAND": ["House of Pain", "Funkdoobiest", "Rage Against the Machine", "Seven Year Bitch"], "MISC": ["Soul Assassins"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MISC, ALBUM\n\nText:\nThe band headlined the Soul Assassins tour with House of Pain and Funkdoobiest as support, then performed on a college tour with Rage Against the Machine and Seven Year Bitch.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MISC, ALBUM\n\nText:\nThe band headlined the Soul Assassins tour with House of Pain and Funkdoobiest as support, then performed on a college tour with Rage Against the Machine and Seven Year Bitch.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_78", "text": "They are run by the British Phonographic Industry and took place on 17 February 2004 at Earls Court Exhibition Centre in London.", "requested_types": ["ALBUM", "AWARD", "LOCATION", "BAND", "ORGANISATION"], "distractors": ["BAND", "ALBUM", "AWARD"], "gold": {"ALBUM": [], "AWARD": [], "LOCATION": ["Earls Court Exhibition Centre", "London"], "BAND": [], "ORGANISATION": ["British Phonographic Industry"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, AWARD, LOCATION, BAND, ORGANISATION\n\nText:\nThey are run by the British Phonographic Industry and took place on 17 February 2004 at Earls Court Exhibition Centre in London.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, AWARD, LOCATION, BAND, ORGANISATION\n\nText:\nThey are run by the British Phonographic Industry and took place on 17 February 2004 at Earls Court Exhibition Centre in London.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_79", "text": "They play Avant-garde jazz versions of tradition American Folk music & amp; Blues songs with Ritchie's shakuhachi playing as the focal point.", "requested_types": ["MUSIC_GENRE", "PERSON", "MUSICAL_INSTRUMENT", "MUSICAL_ARTIST", "LOCATION"], "distractors": ["LOCATION", "PERSON"], "gold": {"MUSIC_GENRE": ["Avant-garde jazz", "American Folk music", "Blues"], "PERSON": [], "MUSICAL_INSTRUMENT": ["shakuhachi"], "MUSICAL_ARTIST": ["Ritchie"], "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: MUSIC_GENRE, PERSON, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, LOCATION\n\nText:\nThey play Avant-garde jazz versions of tradition American Folk music & amp; Blues songs with Ritchie's shakuhachi playing as the focal point.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, PERSON, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, LOCATION\n\nText:\nThey play Avant-garde jazz versions of tradition American Folk music & amp; Blues songs with Ritchie's shakuhachi playing as the focal point.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_80", "text": "He received a Kennedy Center Honors in 2009, a Hollywood Walk of Fame star in 2010, the 41st AFI Life Achievement Award in June 2013, a British Film Institute Fellowship in March 2015, a National Medal of Arts in September 2016, and a BAFTA Fellowship in February 2017.", "requested_types": ["AWARD", "LOCATION", "ALBUM"], "distractors": ["ALBUM"], "gold": {"AWARD": ["Kennedy Center Honors", "41st AFI Life Achievement Award", "British Film Institute Fellowship", "National Medal of Arts", "BAFTA Fellowship"], "LOCATION": ["Hollywood Walk of Fame"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, ALBUM\n\nText:\nHe received a Kennedy Center Honors in 2009, a Hollywood Walk of Fame star in 2010, the 41st AFI Life Achievement Award in June 2013, a British Film Institute Fellowship in March 2015, a National Medal of Arts in September 2016, and a BAFTA Fellowship in February 2017.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, ALBUM\n\nText:\nHe received a Kennedy Center Honors in 2009, a Hollywood Walk of Fame star in 2010, the 41st AFI Life Achievement Award in June 2013, a British Film Institute Fellowship in March 2015, a National Medal of Arts in September 2016, and a BAFTA Fellowship in February 2017.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_81", "text": "Shulman, Art Dynamo - Country Style (1956), TV Guide, p, 28 The late 1950s saw the emergence of Buddy Holly, but by the end of the decade, backlash as well as traditional artists such as Ray Price, Marty Robbins, and Johnny Horton began to shift the industry away from the rock n 'roll influences of the mid-1950s.", "requested_types": ["MUSICAL_INSTRUMENT", "MUSIC_GENRE", "MISC", "PERSON", "SONG", "ALBUM", "MUSICAL_ARTIST", "COUNTRY"], "distractors": ["COUNTRY", "MUSICAL_INSTRUMENT", "SONG"], "gold": {"MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": ["rock n 'roll"], "MISC": ["Dynamo - Country Style"], "PERSON": ["Shulman, Art"], "SONG": [], "ALBUM": ["Buddy Holly"], "MUSICAL_ARTIST": ["Ray Price", "Marty Robbins", "Johnny Horton"], "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: MUSICAL_INSTRUMENT, MUSIC_GENRE, MISC, PERSON, SONG, ALBUM, MUSICAL_ARTIST, COUNTRY\n\nText:\nShulman, Art Dynamo - Country Style (1956), TV Guide, p, 28 The late 1950s saw the emergence of Buddy Holly, but by the end of the decade, backlash as well as traditional artists such as Ray Price, Marty Robbins, and Johnny Horton began to shift the industry away from the rock n 'roll influences of the mid-1950s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSIC_GENRE, MISC, PERSON, SONG, ALBUM, MUSICAL_ARTIST, COUNTRY\n\nText:\nShulman, Art Dynamo - Country Style (1956), TV Guide, p, 28 The late 1950s saw the emergence of Buddy Holly, but by the end of the decade, backlash as well as traditional artists such as Ray Price, Marty Robbins, and Johnny Horton began to shift the industry away from the rock n 'roll influences of the mid-1950s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_82", "text": "By the time she released her fourth album in the early 1990s, she had amassed several top ten singles in the UK and Australia, including I Should Be So Lucky, The Loco-Motion, Hand on Your Heart, Better the Devil You Know and Step Back in Time.", "requested_types": ["COUNTRY", "SONG", "PERSON"], "distractors": ["PERSON"], "gold": {"COUNTRY": ["UK", "Australia"], "SONG": ["I Should Be So Lucky", "The Loco-Motion", "Hand on Your Heart", "Better the Devil You Know", "Step Back in Time"], "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: COUNTRY, SONG, PERSON\n\nText:\nBy the time she released her fourth album in the early 1990s, she had amassed several top ten singles in the UK and Australia, including I Should Be So Lucky, The Loco-Motion, Hand on Your Heart, Better the Devil You Know and Step Back in 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: COUNTRY, SONG, PERSON\n\nText:\nBy the time she released her fourth album in the early 1990s, she had amassed several top ten singles in the UK and Australia, including I Should Be So Lucky, The Loco-Motion, Hand on Your Heart, Better the Devil You Know and Step Back in 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_music_83", "text": "Mathis has undergone rehabilitation for both alcohol and prescription drug addictions, and he has supported many organizations through the years, including the American Cancer Society, the March of Dimes, the YWCA and YMCA, the Muscular Dystrophy Association and the NAACP.", "requested_types": ["MISC", "MUSIC_GENRE", "LOCATION", "ORGANISATION", "MUSICAL_ARTIST"], "distractors": ["LOCATION", "MUSIC_GENRE", "MISC"], "gold": {"MISC": [], "MUSIC_GENRE": [], "LOCATION": [], "ORGANISATION": ["American Cancer Society", "March of Dimes", "YWCA", "YMCA", "Muscular Dystrophy Association", "NAACP"], "MUSICAL_ARTIST": ["Mathis"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, LOCATION, ORGANISATION, MUSICAL_ARTIST\n\nText:\nMathis has undergone rehabilitation for both alcohol and prescription drug addictions, and he has supported many organizations through the years, including the American Cancer Society, the March of Dimes, the YWCA and YMCA, the Muscular Dystrophy Association and the NAACP.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, LOCATION, ORGANISATION, MUSICAL_ARTIST\n\nText:\nMathis has undergone rehabilitation for both alcohol and prescription drug addictions, and he has supported many organizations through the years, including the American Cancer Society, the March of Dimes, the YWCA and YMCA, the Muscular Dystrophy Association and the NAACP.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_84", "text": "She had her first experience in show business when she was crowned Miss Teenager Universal in 1971, and was Miss World / Venezuela in 1975 where she became sixth runner-up in the Miss World pageant won by Puerto Rico's Wilnelia Merced.", "requested_types": ["COUNTRY", "EVENT", "SONG", "MUSICAL_ARTIST", "AWARD", "MUSICAL_INSTRUMENT", "LOCATION", "PERSON"], "distractors": ["MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "SONG"], "gold": {"COUNTRY": ["Venezuela"], "EVENT": ["Miss World pageant"], "SONG": [], "MUSICAL_ARTIST": [], "AWARD": ["Miss Teenager Universal", "Miss World"], "MUSICAL_INSTRUMENT": [], "LOCATION": ["Puerto Rico"], "PERSON": ["Wilnelia Merced"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST, AWARD, MUSICAL_INSTRUMENT, LOCATION, PERSON\n\nText:\nShe had her first experience in show business when she was crowned Miss Teenager Universal in 1971, and was Miss World / Venezuela in 1975 where she became sixth runner-up in the Miss World pageant won by Puerto Rico's Wilnelia Merced.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST, AWARD, MUSICAL_INSTRUMENT, LOCATION, PERSON\n\nText:\nShe had her first experience in show business when she was crowned Miss Teenager Universal in 1971, and was Miss World / Venezuela in 1975 where she became sixth runner-up in the Miss World pageant won by Puerto Rico's Wilnelia Merced.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_85", "text": "Nathan Phillips Square was also a venue for Wheelchair tennis for the 2017 Invictus Games.", "requested_types": ["EVENT", "PERSON", "LOCATION", "MUSICAL_INSTRUMENT", "MISC"], "distractors": ["PERSON", "MUSICAL_INSTRUMENT"], "gold": {"EVENT": ["2017 Invictus Games"], "PERSON": [], "LOCATION": ["Nathan Phillips Square"], "MUSICAL_INSTRUMENT": [], "MISC": ["Wheelchair tennis"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, MUSICAL_INSTRUMENT, MISC\n\nText:\nNathan Phillips Square was also a venue for Wheelchair tennis for the 2017 Invictus 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, PERSON, LOCATION, MUSICAL_INSTRUMENT, MISC\n\nText:\nNathan Phillips Square was also a venue for Wheelchair tennis for the 2017 Invictus 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_music_86", "text": "Based on the songs of the Spice Girls, the show began previews at the Piccadilly Theatre in the West End on 27 November 2012 and had its Press Night on 11 December 2012 and features some of the group's biggest hit songs including Wannabe, Spice Up Your Life and the eponymous Viva Forever.", "requested_types": ["LOCATION", "MISC", "EVENT", "SONG", "BAND", "ORGANISATION", "COUNTRY"], "distractors": ["MISC", "ORGANISATION", "COUNTRY"], "gold": {"LOCATION": ["Piccadilly Theatre"], "MISC": [], "EVENT": ["West End", "Press Night"], "SONG": ["Wannabe", "Spice Up Your Life", "Viva Forever"], "BAND": ["Spice Girls"], "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: LOCATION, MISC, EVENT, SONG, BAND, ORGANISATION, COUNTRY\n\nText:\nBased on the songs of the Spice Girls, the show began previews at the Piccadilly Theatre in the West End on 27 November 2012 and had its Press Night on 11 December 2012 and features some of the group's biggest hit songs including Wannabe, Spice Up Your Life and the eponymous Viva Forever.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT, SONG, BAND, ORGANISATION, COUNTRY\n\nText:\nBased on the songs of the Spice Girls, the show began previews at the Piccadilly Theatre in the West End on 27 November 2012 and had its Press Night on 11 December 2012 and features some of the group's biggest hit songs including Wannabe, Spice Up Your Life and the eponymous Viva Forever.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_87", "text": "The third generation (1950s-1960s) started at the end of World War II with mountaineer string band music known as Bluegrass music, which emerged when Bill Monroe, along with Lester Flatt and Earl Scruggs were introduced by Roy Acuff at the Grand Ole Opry.", "requested_types": ["MUSIC_GENRE", "LOCATION", "AWARD", "MISC", "MUSICAL_ARTIST", "SONG", "EVENT"], "distractors": ["MISC", "SONG", "AWARD"], "gold": {"MUSIC_GENRE": ["Bluegrass music"], "LOCATION": ["Grand Ole Opry"], "AWARD": [], "MISC": [], "MUSICAL_ARTIST": ["Bill Monroe", "Lester Flatt", "Earl Scruggs", "Roy Acuff"], "SONG": [], "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: MUSIC_GENRE, LOCATION, AWARD, MISC, MUSICAL_ARTIST, SONG, EVENT\n\nText:\nThe third generation (1950s-1960s) started at the end of World War II with mountaineer string band music known as Bluegrass music, which emerged when Bill Monroe, along with Lester Flatt and Earl Scruggs were introduced by Roy Acuff at the Grand Ole Opry.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, LOCATION, AWARD, MISC, MUSICAL_ARTIST, SONG, EVENT\n\nText:\nThe third generation (1950s-1960s) started at the end of World War II with mountaineer string band music known as Bluegrass music, which emerged when Bill Monroe, along with Lester Flatt and Earl Scruggs were introduced by Roy Acuff at the Grand Ole Opry.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_88", "text": "Their styles developed into West Coast blues, Detroit blues, and post-World War II Chicago blues, which differed from earlier, predominantly acoustic-style blues.", "requested_types": ["MISC", "SONG", "MUSIC_GENRE"], "distractors": ["SONG"], "gold": {"MISC": ["post-World War II"], "SONG": [], "MUSIC_GENRE": ["West Coast blues", "Detroit blues", "Chicago blues", "acoustic-style blues"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE\n\nText:\nTheir styles developed into West Coast blues, Detroit blues, and post-World War II Chicago blues, which differed from earlier, predominantly acoustic-style blues.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE\n\nText:\nTheir styles developed into West Coast blues, Detroit blues, and post-World War II Chicago blues, which differed from earlier, predominantly acoustic-style blues.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_89", "text": "Just like in Christian rock and other Christian music genres, some artists welcome being called Christian artists while others do not want to be labeled as Christian music, as to not limit their music to the Christian music market.", "requested_types": ["MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "MUSIC_GENRE"], "distractors": ["MUSICAL_ARTIST", "MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_ARTIST": [], "MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": ["Christian rock", "Christian music"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MUSICAL_INSTRUMENT, MUSIC_GENRE\n\nText:\nJust like in Christian rock and other Christian music genres, some artists welcome being called Christian artists while others do not want to be labeled as Christian music, as to not limit their music to the Christian music market.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MUSICAL_INSTRUMENT, MUSIC_GENRE\n\nText:\nJust like in Christian rock and other Christian music genres, some artists welcome being called Christian artists while others do not want to be labeled as Christian music, as to not limit their music to the Christian music market.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_90", "text": "He also co-wrote Posible, which has been used as a theme song for the 2005 Southeast Asian Games.", "requested_types": ["EVENT", "ORGANISATION", "PERSON", "MUSICAL_INSTRUMENT", "SONG"], "distractors": ["MUSICAL_INSTRUMENT", "PERSON", "ORGANISATION"], "gold": {"EVENT": ["2005 Southeast Asian Games"], "ORGANISATION": [], "PERSON": [], "MUSICAL_INSTRUMENT": [], "SONG": ["Posible"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, MUSICAL_INSTRUMENT, SONG\n\nText:\nHe also co-wrote Posible, which has been used as a theme song for the 2005 Southeast 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, ORGANISATION, PERSON, MUSICAL_INSTRUMENT, SONG\n\nText:\nHe also co-wrote Posible, which has been used as a theme song for the 2005 Southeast 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_music_91", "text": "Derived from the traditional Western, including Red Dirt, New Mexico music, Texas country music, Tejano music, and honky-tonk musical styles of the late 1950s and 1960s.", "requested_types": ["PERSON", "MUSIC_GENRE", "LOCATION", "MUSICAL_ARTIST"], "distractors": ["PERSON", "LOCATION", "MUSICAL_ARTIST"], "gold": {"PERSON": [], "MUSIC_GENRE": ["traditional Western", "Red Dirt", "New Mexico music", "Texas country music", "Tejano music", "honky-tonk"], "LOCATION": [], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, LOCATION, MUSICAL_ARTIST\n\nText:\nDerived from the traditional Western, including Red Dirt, New Mexico music, Texas country music, Tejano music, and honky-tonk musical styles of the late 1950s and 1960s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, LOCATION, MUSICAL_ARTIST\n\nText:\nDerived from the traditional Western, including Red Dirt, New Mexico music, Texas country music, Tejano music, and honky-tonk musical styles of the late 1950s and 1960s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_92", "text": "Their farewell concerts at Wembley Arena were multiple sell-outs; their final concert took place at the Brighton Centre on 11 December 1982.", "requested_types": ["PERSON", "BAND", "LOCATION"], "distractors": ["PERSON", "BAND"], "gold": {"PERSON": [], "BAND": [], "LOCATION": ["Wembley Arena", "Brighton Centre"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, LOCATION\n\nText:\nTheir farewell concerts at Wembley Arena were multiple sell-outs; their final concert took place at the Brighton Centre on 11 December 1982.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, LOCATION\n\nText:\nTheir farewell concerts at Wembley Arena were multiple sell-outs; their final concert took place at the Brighton Centre on 11 December 1982.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_93", "text": "It is recognized as a primary instrument in genres such as blues, Bluegrass music, country, flamenco, Folk music, jazz, jota, mariachi, metal, punk, reggae, rock, Soul music, and pop.", "requested_types": ["ALBUM", "SONG", "MUSIC_GENRE", "MUSICAL_INSTRUMENT"], "distractors": ["SONG", "MUSICAL_INSTRUMENT", "ALBUM"], "gold": {"ALBUM": [], "SONG": [], "MUSIC_GENRE": ["blues", "Bluegrass music", "country", "flamenco", "Folk music", "jazz", "jota", "mariachi", "metal", "punk", "reggae", "rock", "Soul music", "pop"], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, SONG, MUSIC_GENRE, MUSICAL_INSTRUMENT\n\nText:\nIt is recognized as a primary instrument in genres such as blues, Bluegrass music, country, flamenco, Folk music, jazz, jota, mariachi, metal, punk, reggae, rock, Soul music, and pop.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, SONG, MUSIC_GENRE, MUSICAL_INSTRUMENT\n\nText:\nIt is recognized as a primary instrument in genres such as blues, Bluegrass music, country, flamenco, Folk music, jazz, jota, mariachi, metal, punk, reggae, rock, Soul music, and pop.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_94", "text": "Parsons acted as Assistant Engineer on the Beatles' albums Abbey Road (1969) and Let It Be (1970), engineered Pink Floyd 's The Dark Side of the Moon (1973), and produced several acts for EMI Records.", "requested_types": ["MUSICAL_INSTRUMENT", "BAND", "ALBUM", "EVENT", "LOCATION", "PERSON", "ORGANISATION"], "distractors": ["MUSICAL_INSTRUMENT", "EVENT", "LOCATION"], "gold": {"MUSICAL_INSTRUMENT": [], "BAND": ["the Beatles", "Pink Floyd"], "ALBUM": ["Abbey Road", "Let It Be", "The Dark Side of the Moon"], "EVENT": [], "LOCATION": [], "PERSON": ["Parsons"], "ORGANISATION": ["EMI Records"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, BAND, ALBUM, EVENT, LOCATION, PERSON, ORGANISATION\n\nText:\nParsons acted as Assistant Engineer on the Beatles' albums Abbey Road (1969) and Let It Be (1970), engineered Pink Floyd 's The Dark Side of the Moon (1973), and produced several acts for EMI Records.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, BAND, ALBUM, EVENT, LOCATION, PERSON, ORGANISATION\n\nText:\nParsons acted as Assistant Engineer on the Beatles' albums Abbey Road (1969) and Let It Be (1970), engineered Pink Floyd 's The Dark Side of the Moon (1973), and produced several acts for EMI Records.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_95", "text": "The West African community is also very large, integrated by people from Senegal, Mali, Ivory Coast, and Guinea.", "requested_types": ["SONG", "PERSON", "COUNTRY", "LOCATION", "ORGANISATION"], "distractors": ["SONG", "ORGANISATION", "PERSON"], "gold": {"SONG": [], "PERSON": [], "COUNTRY": ["Senegal", "Mali", "Ivory Coast", "Guinea"], "LOCATION": ["West African"], "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: SONG, PERSON, COUNTRY, LOCATION, ORGANISATION\n\nText:\nThe West African community is also very large, integrated by people from Senegal, Mali, Ivory Coast, and Guinea.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, PERSON, COUNTRY, LOCATION, ORGANISATION\n\nText:\nThe West African community is also very large, integrated by people from Senegal, Mali, Ivory Coast, and Guinea.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_96", "text": "On February 17, 2020, a 50th anniversary concert production of Joseph was staged at Lincoln Center 's David Geffen Hall in New York City.", "requested_types": ["MUSIC_GENRE", "LOCATION", "ORGANISATION"], "distractors": ["MUSIC_GENRE"], "gold": {"MUSIC_GENRE": [], "LOCATION": ["Lincoln Center", "David Geffen Hall", "New York City"], "ORGANISATION": ["Joseph"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, LOCATION, ORGANISATION\n\nText:\nOn February 17, 2020, a 50th anniversary concert production of Joseph was staged at Lincoln Center 's David Geffen Hall in New York 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: MUSIC_GENRE, LOCATION, ORGANISATION\n\nText:\nOn February 17, 2020, a 50th anniversary concert production of Joseph was staged at Lincoln Center 's David Geffen Hall in New York 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_music_97", "text": "In parts of Europe, including Germany, the Czech Republic, Slovakia, Poland, Hungary, Norway, Denmark, Serbia, Croatia, Slovenia, Finland and Iceland (and Sweden before about 1990s), the Gothic b transformed into the letter H (possibly for hart, German for hard, or just because the Gothic b resembled an H).", "requested_types": ["ORGANISATION", "COUNTRY", "MUSIC_GENRE", "PERSON", "BAND", "ALBUM", "LOCATION"], "distractors": ["ORGANISATION", "PERSON", "BAND"], "gold": {"ORGANISATION": [], "COUNTRY": ["Germany", "Czech Republic", "Slovakia", "Poland", "Hungary", "Norway", "Serbia", "Croatia", "Slovenia", "Finland", "Iceland", "Sweden", "German"], "MUSIC_GENRE": ["Gothic"], "PERSON": [], "BAND": [], "ALBUM": ["Denmark"], "LOCATION": ["Europe"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, PERSON, BAND, ALBUM, LOCATION\n\nText:\nIn parts of Europe, including Germany, the Czech Republic, Slovakia, Poland, Hungary, Norway, Denmark, Serbia, Croatia, Slovenia, Finland and Iceland (and Sweden before about 1990s), the Gothic b transformed into the letter H (possibly for hart, German for hard, or just because the Gothic b resembled an 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: ORGANISATION, COUNTRY, MUSIC_GENRE, PERSON, BAND, ALBUM, LOCATION\n\nText:\nIn parts of Europe, including Germany, the Czech Republic, Slovakia, Poland, Hungary, Norway, Denmark, Serbia, Croatia, Slovenia, Finland and Iceland (and Sweden before about 1990s), the Gothic b transformed into the letter H (possibly for hart, German for hard, or just because the Gothic b resembled an 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_music_98", "text": "Their musical style shifted in the direction of progressive rock with the albums Aqualung (1971), Thick as a Brick (1972) and A Passion Play (1973), and shifted again to hard rock mixed with folk rock with Songs from the Wood (1977) and Heavy Horses (1978).", "requested_types": ["MUSIC_GENRE", "MISC", "SONG", "ALBUM"], "distractors": ["SONG", "MISC"], "gold": {"MUSIC_GENRE": ["progressive rock", "hard rock", "folk rock"], "MISC": [], "SONG": [], "ALBUM": ["Aqualung", "Thick as a Brick", "A Passion Play", "Songs from the Wood", "Heavy Horses"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MISC, SONG, ALBUM\n\nText:\nTheir musical style shifted in the direction of progressive rock with the albums Aqualung (1971), Thick as a Brick (1972) and A Passion Play (1973), and shifted again to hard rock mixed with folk rock with Songs from the Wood (1977) and Heavy Horses (1978).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MISC, SONG, ALBUM\n\nText:\nTheir musical style shifted in the direction of progressive rock with the albums Aqualung (1971), Thick as a Brick (1972) and A Passion Play (1973), and shifted again to hard rock mixed with folk rock with Songs from the Wood (1977) and Heavy Horses (1978).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_99", "text": "Examples include Portuguese Fado, Spanish-speaking Mexican Regional, Reggaeton and Tejano music, French Cajun music (especially in French Louisiana), Russian Shanson, and (since the late 2000s) Korean K-pop.", "requested_types": ["LOCATION", "MUSICAL_INSTRUMENT", "MISC", "COUNTRY", "MUSIC_GENRE", "ALBUM"], "distractors": ["ALBUM", "COUNTRY", "MUSICAL_INSTRUMENT"], "gold": {"LOCATION": ["French Louisiana"], "MUSICAL_INSTRUMENT": [], "MISC": ["Portuguese", "French", "Russian", "Korean"], "COUNTRY": [], "MUSIC_GENRE": ["Fado", "Mexican Regional", "Reggaeton", "Tejano music", "Cajun music", "Shanson", "K-pop"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, MISC, COUNTRY, MUSIC_GENRE, ALBUM\n\nText:\nExamples include Portuguese Fado, Spanish-speaking Mexican Regional, Reggaeton and Tejano music, French Cajun music (especially in French Louisiana), Russian Shanson, and (since the late 2000s) Korean K-pop.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, MISC, COUNTRY, MUSIC_GENRE, ALBUM\n\nText:\nExamples include Portuguese Fado, Spanish-speaking Mexican Regional, Reggaeton and Tejano music, French Cajun music (especially in French Louisiana), Russian Shanson, and (since the late 2000s) Korean K-pop.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_100", "text": "Shortly afterwards, her career in the United States was launched with a concert at New York's Lincoln Center, followed in 1973 by the first of many Carnegie Hall appearances.", "requested_types": ["MUSIC_GENRE", "LOCATION", "ORGANISATION", "COUNTRY"], "distractors": ["ORGANISATION", "MUSIC_GENRE"], "gold": {"MUSIC_GENRE": [], "LOCATION": ["New York", "Lincoln Center", "Carnegie Hall"], "ORGANISATION": [], "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: MUSIC_GENRE, LOCATION, ORGANISATION, COUNTRY\n\nText:\nShortly afterwards, her career in the United States was launched with a concert at New York's Lincoln Center, followed in 1973 by the first of many Carnegie Hall appearances.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, LOCATION, ORGANISATION, COUNTRY\n\nText:\nShortly afterwards, her career in the United States was launched with a concert at New York's Lincoln Center, followed in 1973 by the first of many Carnegie Hall appearances.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_101", "text": "The Godfather won three Academy Awards: Academy Award for Best Picture, Academy Award for Best Actor, and Academy Award for Best Adapted Screenplay (shared with Mario Puzo).", "requested_types": ["MUSIC_GENRE", "PERSON", "BAND", "MISC", "AWARD"], "distractors": ["MISC", "MUSIC_GENRE"], "gold": {"MUSIC_GENRE": [], "PERSON": ["Mario Puzo"], "BAND": ["The Godfather"], "MISC": [], "AWARD": ["Academy Awards", "Academy Award for Best Picture", "Academy Award for Best Actor", "Academy Award for Best Adapted Screenplay"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, PERSON, BAND, MISC, AWARD\n\nText:\nThe Godfather won three Academy Awards: Academy Award for Best Picture, Academy Award for Best Actor, and Academy Award for Best Adapted Screenplay (shared with Mario Puzo).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, PERSON, BAND, MISC, AWARD\n\nText:\nThe Godfather won three Academy Awards: Academy Award for Best Picture, Academy Award for Best Actor, and Academy Award for Best Adapted Screenplay (shared with Mario Puzo).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_102", "text": "The banned monastic orders: Jesuits, Camaldolese, Order of Friars Minor Capuchin, Carmelites, Carthusians, Poor Clares, Order of Saint Benedict, Cistercians, Dominican Order (Order of Preachers), Franciscans, Order of Saint Paul the First Hermit and Premonstratensians, and their wealth was taken over by the Religious Fund.", "requested_types": ["MUSICAL_ARTIST", "ORGANISATION", "MUSICAL_INSTRUMENT", "AWARD"], "distractors": ["MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "AWARD"], "gold": {"MUSICAL_ARTIST": [], "ORGANISATION": ["Jesuits", "Camaldolese", "Order of Friars Minor Capuchin", "Carmelites", "Carthusians", "Poor Clares", "Order of Saint Benedict", "Cistercians", "Dominican Order", "Order of Preachers", "Franciscans", "Order of Saint Paul the First Hermit", "Premonstratensians", "Religious Fund"], "MUSICAL_INSTRUMENT": [], "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: MUSICAL_ARTIST, ORGANISATION, MUSICAL_INSTRUMENT, AWARD\n\nText:\nThe banned monastic orders: Jesuits, Camaldolese, Order of Friars Minor Capuchin, Carmelites, Carthusians, Poor Clares, Order of Saint Benedict, Cistercians, Dominican Order (Order of Preachers), Franciscans, Order of Saint Paul the First Hermit and Premonstratensians, and their wealth was taken over by the Religious 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: MUSICAL_ARTIST, ORGANISATION, MUSICAL_INSTRUMENT, AWARD\n\nText:\nThe banned monastic orders: Jesuits, Camaldolese, Order of Friars Minor Capuchin, Carmelites, Carthusians, Poor Clares, Order of Saint Benedict, Cistercians, Dominican Order (Order of Preachers), Franciscans, Order of Saint Paul the First Hermit and Premonstratensians, and their wealth was taken over by the Religious 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_music_103", "text": "Black metal album covers are typically dark and tend to be atmospheric or provocative; some feature natural or fantasy landscapes (for example Burzum 's Filosofem and Emperor's In the Nightside Eclipse) while others are violent, sexually transgressive, sacrilegious, or iconoclastic (for example Marduk' s Fuck Me Jesus and Dimmu Borgir 's In Sorte Diaboli).", "requested_types": ["MUSIC_GENRE", "AWARD", "ALBUM", "BAND", "MUSICAL_INSTRUMENT", "MISC"], "distractors": ["MISC", "MUSICAL_INSTRUMENT", "AWARD"], "gold": {"MUSIC_GENRE": ["Black metal"], "AWARD": [], "ALBUM": ["Filosofem", "In the Nightside Eclipse", "Fuck Me Jesus", "In Sorte Diaboli"], "BAND": ["Burzum", "Emperor", "Marduk", "Dimmu Borgir"], "MUSICAL_INSTRUMENT": [], "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: MUSIC_GENRE, AWARD, ALBUM, BAND, MUSICAL_INSTRUMENT, MISC\n\nText:\nBlack metal album covers are typically dark and tend to be atmospheric or provocative; some feature natural or fantasy landscapes (for example Burzum 's Filosofem and Emperor's In the Nightside Eclipse) while others are violent, sexually transgressive, sacrilegious, or iconoclastic (for example Marduk' s Fuck Me Jesus and Dimmu Borgir 's In Sorte Diaboli).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, AWARD, ALBUM, BAND, MUSICAL_INSTRUMENT, MISC\n\nText:\nBlack metal album covers are typically dark and tend to be atmospheric or provocative; some feature natural or fantasy landscapes (for example Burzum 's Filosofem and Emperor's In the Nightside Eclipse) while others are violent, sexually transgressive, sacrilegious, or iconoclastic (for example Marduk' s Fuck Me Jesus and Dimmu Borgir 's In Sorte Diaboli).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_104", "text": "Local varieties of this dance are also found in the Nordic countries, United Kingdom, Republic of Ireland, Latin America (especially Mexico), and in the United States.", "requested_types": ["LOCATION", "COUNTRY", "EVENT", "ALBUM", "MISC", "MUSIC_GENRE"], "distractors": ["EVENT", "MUSIC_GENRE", "ALBUM"], "gold": {"LOCATION": ["Latin America"], "COUNTRY": ["United Kingdom", "Republic of Ireland", "Mexico", "United States"], "EVENT": [], "ALBUM": [], "MISC": ["Nordic countries"], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, EVENT, ALBUM, MISC, MUSIC_GENRE\n\nText:\nLocal varieties of this dance are also found in the Nordic countries, United Kingdom, Republic of Ireland, Latin America (especially Mexico), and 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: LOCATION, COUNTRY, EVENT, ALBUM, MISC, MUSIC_GENRE\n\nText:\nLocal varieties of this dance are also found in the Nordic countries, United Kingdom, Republic of Ireland, Latin America (especially Mexico), and 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_music_105", "text": "The Dark Star St. Stephen pairing was taken from the February 27, 1969 show at the The Fillmore; The Eleven and Turn On Your Love Light were from the January 26, 1969 show at the Avalon Ballroom; Death Don't Have No Mercy, Feedback, and And We Bid You Goodnight were recorded March 2, 1969, at the The Fillmore.", "requested_types": ["SONG", "COUNTRY", "BAND", "LOCATION"], "distractors": ["BAND", "COUNTRY"], "gold": {"SONG": ["Dark Star St. Stephen", "The Eleven", "Turn On Your Love Light", "Death Don't Have No Mercy", "Feedback", "We Bid You Goodnight"], "COUNTRY": [], "BAND": [], "LOCATION": ["The Fillmore", "Avalon Ballroom"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, COUNTRY, BAND, LOCATION\n\nText:\nThe Dark Star St. Stephen pairing was taken from the February 27, 1969 show at the The Fillmore; The Eleven and Turn On Your Love Light were from the January 26, 1969 show at the Avalon Ballroom; Death Don't Have No Mercy, Feedback, and And We Bid You Goodnight were recorded March 2, 1969, at the The Fillmore.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, COUNTRY, BAND, LOCATION\n\nText:\nThe Dark Star St. Stephen pairing was taken from the February 27, 1969 show at the The Fillmore; The Eleven and Turn On Your Love Light were from the January 26, 1969 show at the Avalon Ballroom; Death Don't Have No Mercy, Feedback, and And We Bid You Goodnight were recorded March 2, 1969, at the The Fillmore.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_106", "text": "McTell's influence extended over a wide variety of artists, including the The Allman Brothers Band, who covered his Statesboro Blues, and Bob Dylan, who paid tribute to him in his 1983 song Blind Willie McTell, the refrain of which is And I know no one can sing the blues like Blind Willie McTell.", "requested_types": ["EVENT", "BAND", "MUSIC_GENRE", "MUSICAL_ARTIST", "PERSON", "MISC", "SONG"], "distractors": ["MISC", "MUSIC_GENRE", "EVENT"], "gold": {"EVENT": [], "BAND": ["The Allman Brothers Band"], "MUSIC_GENRE": [], "MUSICAL_ARTIST": ["McTell", "Blind Willie McTell"], "PERSON": ["Bob Dylan"], "MISC": [], "SONG": ["Statesboro Blues"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSIC_GENRE, MUSICAL_ARTIST, PERSON, MISC, SONG\n\nText:\nMcTell's influence extended over a wide variety of artists, including the The Allman Brothers Band, who covered his Statesboro Blues, and Bob Dylan, who paid tribute to him in his 1983 song Blind Willie McTell, the refrain of which is And I know no one can sing the blues like Blind Willie McTell.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSIC_GENRE, MUSICAL_ARTIST, PERSON, MISC, SONG\n\nText:\nMcTell's influence extended over a wide variety of artists, including the The Allman Brothers Band, who covered his Statesboro Blues, and Bob Dylan, who paid tribute to him in his 1983 song Blind Willie McTell, the refrain of which is And I know no one can sing the blues like Blind Willie McTell.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_107", "text": "He also performed in The Rocky Horror Show, as the narrator, at the Churchill Theatre in Bromley and the New Wimbledon Theatre.", "requested_types": ["MISC", "LOCATION", "ALBUM"], "distractors": ["ALBUM"], "gold": {"MISC": ["The Rocky Horror Show"], "LOCATION": ["Churchill Theatre", "Bromley", "New Wimbledon Theatre"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nHe also performed in The Rocky Horror Show, as the narrator, at the Churchill Theatre in Bromley and the New Wimbledon Theatre.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nHe also performed in The Rocky Horror Show, as the narrator, at the Churchill Theatre in Bromley and the New Wimbledon Theatre.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_108", "text": "Alt-country, in various iterations overlapped with other genres, including Red Dirt country music (Cross Canadian Ragweed), jam band s (My Morning Jacket and The String Cheese Incident), and indie folk (The Avett Brothers).", "requested_types": ["BAND", "LOCATION", "MUSIC_GENRE"], "distractors": ["LOCATION"], "gold": {"BAND": ["Cross Canadian Ragweed", "jam band", "My Morning Jacket", "The String Cheese Incident", "The Avett Brothers"], "LOCATION": [], "MUSIC_GENRE": ["Alt-country", "Red Dirt country music", "indie folk"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, LOCATION, MUSIC_GENRE\n\nText:\nAlt-country, in various iterations overlapped with other genres, including Red Dirt country music (Cross Canadian Ragweed), jam band s (My Morning Jacket and The String Cheese Incident), and indie folk (The Avett Brothers).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, LOCATION, MUSIC_GENRE\n\nText:\nAlt-country, in various iterations overlapped with other genres, including Red Dirt country music (Cross Canadian Ragweed), jam band s (My Morning Jacket and The String Cheese Incident), and indie folk (The Avett Brothers).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_109", "text": "Paul Gilbert composes music in a wide variety of styles, including Pop music, Rock and roll, metal, blues, and funk.", "requested_types": ["MUSIC_GENRE", "MUSICAL_ARTIST", "PERSON"], "distractors": ["PERSON"], "gold": {"MUSIC_GENRE": ["Pop music", "Rock and roll", "metal", "blues", "funk"], "MUSICAL_ARTIST": ["Paul Gilbert"], "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: MUSIC_GENRE, MUSICAL_ARTIST, PERSON\n\nText:\nPaul Gilbert composes music in a wide variety of styles, including Pop music, Rock and roll, metal, blues, and funk.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, PERSON\n\nText:\nPaul Gilbert composes music in a wide variety of styles, including Pop music, Rock and roll, metal, blues, and funk.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_110", "text": "The follow-up release, Road Rage, reached number 5 in May, and was nominated for best song at the Brit Awards and the Ivor Novello Awards, winning at the Q Awards.", "requested_types": ["LOCATION", "ORGANISATION", "AWARD", "SONG"], "distractors": ["ORGANISATION", "LOCATION"], "gold": {"LOCATION": [], "ORGANISATION": [], "AWARD": ["Brit Awards", "Ivor Novello Awards", "Q Awards"], "SONG": ["Road Rage"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, AWARD, SONG\n\nText:\nThe follow-up release, Road Rage, reached number 5 in May, and was nominated for best song at the Brit Awards and the Ivor Novello Awards, winning at the Q 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: LOCATION, ORGANISATION, AWARD, SONG\n\nText:\nThe follow-up release, Road Rage, reached number 5 in May, and was nominated for best song at the Brit Awards and the Ivor Novello Awards, winning at the Q 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_music_111", "text": "With the song's success, the age of rock music began overnight and ended the dominance of the jazz and pop standards performed by Frank Sinatra, Jo Stafford, Perry Como, Bing Crosby, Eddie Fisher, Patti Page and others.", "requested_types": ["ORGANISATION", "MISC", "AWARD", "MUSIC_GENRE", "MUSICAL_ARTIST"], "distractors": ["AWARD", "MISC", "ORGANISATION"], "gold": {"ORGANISATION": [], "MISC": [], "AWARD": [], "MUSIC_GENRE": ["rock music", "jazz", "pop"], "MUSICAL_ARTIST": ["Frank Sinatra", "Jo Stafford", "Perry Como", "Bing Crosby", "Eddie Fisher", "Patti Page"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nWith the song's success, the age of rock music began overnight and ended the dominance of the jazz and pop standards performed by Frank Sinatra, Jo Stafford, Perry Como, Bing Crosby, Eddie Fisher, Patti Page 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: ORGANISATION, MISC, AWARD, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nWith the song's success, the age of rock music began overnight and ended the dominance of the jazz and pop standards performed by Frank Sinatra, Jo Stafford, Perry Como, Bing Crosby, Eddie Fisher, Patti Page 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_music_112", "text": "Govere first attracted media attention by being the first black female gymnast to represent Zimbabwe in artistic gymnastics at the 1999 All-Africa Games, and later by co-founding the Kijana Project, which provides relief for AIDS orphans.", "requested_types": ["ORGANISATION", "EVENT", "PERSON", "LOCATION", "COUNTRY", "MISC"], "distractors": ["LOCATION"], "gold": {"ORGANISATION": ["Kijana Project"], "EVENT": ["1999 All-Africa Games"], "PERSON": ["Govere"], "LOCATION": [], "COUNTRY": ["Zimbabwe"], "MISC": ["AIDS"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, LOCATION, COUNTRY, MISC\n\nText:\nGovere first attracted media attention by being the first black female gymnast to represent Zimbabwe in artistic gymnastics at the 1999 All-Africa Games, and later by co-founding the Kijana Project, which provides relief for AIDS orphans.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, LOCATION, COUNTRY, MISC\n\nText:\nGovere first attracted media attention by being the first black female gymnast to represent Zimbabwe in artistic gymnastics at the 1999 All-Africa Games, and later by co-founding the Kijana Project, which provides relief for AIDS orphans.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_113", "text": "Examples can be found in the music of the Chambers Brothers, George Clinton with his Parliament-Funkadelic collective, Sly and the Family Stone and the productions of Norman Whitfield with The Temptations.", "requested_types": ["LOCATION", "BAND", "MUSIC_GENRE", "MUSICAL_ARTIST"], "distractors": ["LOCATION", "MUSIC_GENRE"], "gold": {"LOCATION": [], "BAND": ["the Chambers Brothers", "Parliament-Funkadelic", "Sly and the Family Stone", "The Temptations"], "MUSIC_GENRE": [], "MUSICAL_ARTIST": ["George Clinton", "Norman Whitfield"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nExamples can be found in the music of the Chambers Brothers, George Clinton with his Parliament-Funkadelic collective, Sly and the Family Stone and the productions of Norman Whitfield with The Temptations.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nExamples can be found in the music of the Chambers Brothers, George Clinton with his Parliament-Funkadelic collective, Sly and the Family Stone and the productions of Norman Whitfield with The Temptations.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_114", "text": "He composed and wrote many standard blues songs of the period, such as Hoochie Coochie Man, I Just Want to Make Love to You (both penned for Muddy Waters) and, Wang Dang Doodle and Back Door Man for Howlin 'Wolf.", "requested_types": ["SONG", "PERSON", "MUSICAL_ARTIST", "LOCATION", "MUSIC_GENRE"], "distractors": ["LOCATION", "PERSON"], "gold": {"SONG": ["Hoochie Coochie Man", "I Just Want to Make Love to You", "Wang Dang Doodle", "Back Door Man"], "PERSON": [], "MUSICAL_ARTIST": ["Muddy Waters", "Howlin 'Wolf"], "LOCATION": [], "MUSIC_GENRE": ["standard blues"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, PERSON, MUSICAL_ARTIST, LOCATION, MUSIC_GENRE\n\nText:\nHe composed and wrote many standard blues songs of the period, such as Hoochie Coochie Man, I Just Want to Make Love to You (both penned for Muddy Waters) and, Wang Dang Doodle and Back Door Man for Howlin 'Wolf.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, PERSON, MUSICAL_ARTIST, LOCATION, MUSIC_GENRE\n\nText:\nHe composed and wrote many standard blues songs of the period, such as Hoochie Coochie Man, I Just Want to Make Love to You (both penned for Muddy Waters) and, Wang Dang Doodle and Back Door Man for Howlin 'Wolf.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_115", "text": "Looking to depart from the distorted production of their previous record, The Downward Spiral (1994), the album features elements of Ambient music and Electronic music music, alongside the band's traditional industrial rock sound.", "requested_types": ["MUSIC_GENRE", "SONG", "ALBUM", "MISC", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_ARTIST", "SONG", "MISC"], "gold": {"MUSIC_GENRE": ["Ambient music", "Electronic music", "industrial rock"], "SONG": [], "ALBUM": ["The Downward Spiral"], "MISC": [], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, SONG, ALBUM, MISC, MUSICAL_ARTIST\n\nText:\nLooking to depart from the distorted production of their previous record, The Downward Spiral (1994), the album features elements of Ambient music and Electronic music music, alongside the band's traditional industrial rock sound.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, SONG, ALBUM, MISC, MUSICAL_ARTIST\n\nText:\nLooking to depart from the distorted production of their previous record, The Downward Spiral (1994), the album features elements of Ambient music and Electronic music music, alongside the band's traditional industrial rock sound.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_116", "text": "G-Mex was also the 2002 Commonwealth Games venue for gymnastics, weightlifting, judo and wrestling.", "requested_types": ["ALBUM", "BAND", "EVENT", "LOCATION", "MISC"], "distractors": ["BAND", "ALBUM"], "gold": {"ALBUM": [], "BAND": [], "EVENT": ["2002 Commonwealth Games"], "LOCATION": ["G-Mex"], "MISC": ["gymnastics", "weightlifting", "judo", "wrestling"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, BAND, EVENT, LOCATION, MISC\n\nText:\nG-Mex was also the 2002 Commonwealth Games venue for gymnastics, weightlifting, judo and wrestling.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, BAND, EVENT, LOCATION, MISC\n\nText:\nG-Mex was also the 2002 Commonwealth Games venue for gymnastics, weightlifting, judo and wrestling.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_117", "text": "Buckingham's Second Hand News, Nicks's Gold Dust Woman and The Chain (the only song written by all five band members) also received significant radio airplay.", "requested_types": ["MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "SONG", "AWARD", "PERSON"], "distractors": ["AWARD", "PERSON", "MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_ARTIST": ["Buckingham", "Nicks"], "MUSICAL_INSTRUMENT": [], "SONG": ["Second Hand News", "Gold Dust Woman", "The Chain"], "AWARD": [], "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: MUSICAL_ARTIST, MUSICAL_INSTRUMENT, SONG, AWARD, PERSON\n\nText:\nBuckingham's Second Hand News, Nicks's Gold Dust Woman and The Chain (the only song written by all five band members) also received significant radio airplay.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MUSICAL_INSTRUMENT, SONG, AWARD, PERSON\n\nText:\nBuckingham's Second Hand News, Nicks's Gold Dust Woman and The Chain (the only song written by all five band members) also received significant radio airplay.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_118", "text": "Johnston was involved in the 1999 All-Africa Games, and in South Africa's bid for the 2004 Summer Olympics.", "requested_types": ["EVENT", "PERSON", "COUNTRY", "MUSICAL_INSTRUMENT", "LOCATION"], "distractors": ["MUSICAL_INSTRUMENT", "LOCATION"], "gold": {"EVENT": ["1999 All-Africa Games", "2004 Summer Olympics"], "PERSON": ["Johnston"], "COUNTRY": ["South Africa"], "MUSICAL_INSTRUMENT": [], "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, PERSON, COUNTRY, MUSICAL_INSTRUMENT, LOCATION\n\nText:\nJohnston was involved in the 1999 All-Africa Games, and in South Africa's bid for the 2004 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: EVENT, PERSON, COUNTRY, MUSICAL_INSTRUMENT, LOCATION\n\nText:\nJohnston was involved in the 1999 All-Africa Games, and in South Africa's bid for the 2004 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_music_119", "text": "Dion's music has been influenced by genres ranging from rock and Rhythm and blues to Gospel music and classical.", "requested_types": ["PERSON", "MUSICAL_INSTRUMENT", "MUSIC_GENRE", "MUSICAL_ARTIST"], "distractors": ["PERSON", "MUSICAL_INSTRUMENT"], "gold": {"PERSON": [], "MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": ["Rhythm and blues", "Gospel music", "classical"], "MUSICAL_ARTIST": ["Dion"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nDion's music has been influenced by genres ranging from rock and Rhythm and blues to Gospel music and classical.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nDion's music has been influenced by genres ranging from rock and Rhythm and blues to Gospel music and classical.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_120", "text": "His songs for the group included Taxman, Within You Without You, While My Guitar Gently Weeps, Here Comes the Sun and Something.", "requested_types": ["COUNTRY", "SONG"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "SONG": ["Taxman", "Within You Without You", "While My Guitar Gently Weeps", "Here Comes the Sun", "Something"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG\n\nText:\nHis songs for the group included Taxman, Within You Without You, While My Guitar Gently Weeps, Here Comes the Sun and Something.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG\n\nText:\nHis songs for the group included Taxman, Within You Without You, While My Guitar Gently Weeps, Here Comes the Sun and Something.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_121", "text": "New songwriting and recording sessions took place, and during October and December, they released the singles The Day Before You Came / Cassandra and Under Attack / You Owe Me One, the A-sides of which were included on the compilation album.", "requested_types": ["MISC", "MUSICAL_ARTIST", "SONG"], "distractors": ["MUSICAL_ARTIST"], "gold": {"MISC": ["A-sides", "compilation album"], "MUSICAL_ARTIST": [], "SONG": ["The Day Before You Came", "Cassandra", "Under Attack", "You Owe Me One"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, SONG\n\nText:\nNew songwriting and recording sessions took place, and during October and December, they released the singles The Day Before You Came / Cassandra and Under Attack / You Owe Me One, the A-sides of which were included on the compilation album.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, SONG\n\nText:\nNew songwriting and recording sessions took place, and during October and December, they released the singles The Day Before You Came / Cassandra and Under Attack / You Owe Me One, the A-sides of which were included on the compilation album.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_122", "text": "John Zorn (born September 2, 1953) is an American composer, arranger, record producer, saxophonist, and multi-instrumentalist with hundreds of album credits as performer, composer, and producer across a variety of genres including jazz, rock, Hardcore punk, classical, Surf music, Heavy metal music, soundtrack, Ambient music, and improvised music.", "requested_types": ["AWARD", "MUSIC_GENRE", "MISC", "MUSICAL_ARTIST"], "distractors": ["AWARD"], "gold": {"AWARD": [], "MUSIC_GENRE": ["jazz", "rock", "Hardcore punk", "classical", "Surf music", "Heavy metal music", "soundtrack", "Ambient music", "improvised music"], "MISC": ["American"], "MUSICAL_ARTIST": ["John Zorn"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, MISC, MUSICAL_ARTIST\n\nText:\nJohn Zorn (born September 2, 1953) is an American composer, arranger, record producer, saxophonist, and multi-instrumentalist with hundreds of album credits as performer, composer, and producer across a variety of genres including jazz, rock, Hardcore punk, classical, Surf music, Heavy metal music, soundtrack, Ambient music, and improvised music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, MISC, MUSICAL_ARTIST\n\nText:\nJohn Zorn (born September 2, 1953) is an American composer, arranger, record producer, saxophonist, and multi-instrumentalist with hundreds of album credits as performer, composer, and producer across a variety of genres including jazz, rock, Hardcore punk, classical, Surf music, Heavy metal music, soundtrack, Ambient music, and improvised music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_123", "text": "For the 2000 World Junior Championships in Athletics, the installation of individual seats was required, which reduced capacity to 66,000 spectators.", "requested_types": ["LOCATION", "MUSIC_GENRE", "EVENT"], "distractors": ["LOCATION", "MUSIC_GENRE"], "gold": {"LOCATION": [], "MUSIC_GENRE": [], "EVENT": ["2000 World Junior 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: LOCATION, MUSIC_GENRE, EVENT\n\nText:\nFor the 2000 World Junior Championships in Athletics, the installation of individual seats was required, which reduced capacity to 66,000 spectators.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, EVENT\n\nText:\nFor the 2000 World Junior Championships in Athletics, the installation of individual seats was required, which reduced capacity to 66,000 spectators.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_124", "text": "Coolio was dropped from Tommy Boy Records and his albums since then, 2001's Coolio.com, 2003's El Cool Magnifico, 2006's The Return of the Gangsta, and 2008's Steal Hear, have not charted on any Billboard chart.", "requested_types": ["MISC", "ORGANISATION", "ALBUM", "MUSICAL_ARTIST", "COUNTRY", "SONG"], "distractors": ["COUNTRY", "SONG"], "gold": {"MISC": ["Billboard"], "ORGANISATION": ["Tommy Boy Records"], "ALBUM": ["Coolio.com", "El Cool Magnifico", "The Return of the Gangsta", "Steal Hear"], "MUSICAL_ARTIST": ["Coolio"], "COUNTRY": [], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MUSICAL_ARTIST, COUNTRY, SONG\n\nText:\nCoolio was dropped from Tommy Boy Records and his albums since then, 2001's Coolio.com, 2003's El Cool Magnifico, 2006's The Return of the Gangsta, and 2008's Steal Hear, have not charted on any Billboard chart.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MUSICAL_ARTIST, COUNTRY, SONG\n\nText:\nCoolio was dropped from Tommy Boy Records and his albums since then, 2001's Coolio.com, 2003's El Cool Magnifico, 2006's The Return of the Gangsta, and 2008's Steal Hear, have not charted on any Billboard chart.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_125", "text": "His complex and intricate fingerpicking inspired Reverend Gary Davis, Jorma Kaukonen, Ry Cooder, Arlen Roth, John Fahey, Ralph McTell, David Bromberg, Leon Redbone and many others.", "requested_types": ["MUSICAL_INSTRUMENT", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_INSTRUMENT": [], "MUSICAL_ARTIST": ["Reverend Gary Davis", "Jorma Kaukonen", "Ry Cooder", "Arlen Roth", "John Fahey", "Ralph McTell", "David Bromberg", "Leon Redbone"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSICAL_ARTIST\n\nText:\nHis complex and intricate fingerpicking inspired Reverend Gary Davis, Jorma Kaukonen, Ry Cooder, Arlen Roth, John Fahey, Ralph McTell, David Bromberg, Leon Redbone and many 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: MUSICAL_INSTRUMENT, MUSICAL_ARTIST\n\nText:\nHis complex and intricate fingerpicking inspired Reverend Gary Davis, Jorma Kaukonen, Ry Cooder, Arlen Roth, John Fahey, Ralph McTell, David Bromberg, Leon Redbone and many 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_music_126", "text": "Due to shared cultural heritage and language, Indian music and Bollywood films are also popular in Afghanistan, Pakistan, Bangladesh, and Nepal, where Hindustani is widely understood.", "requested_types": ["COUNTRY", "MUSICAL_INSTRUMENT", "BAND", "MISC"], "distractors": ["MUSICAL_INSTRUMENT", "BAND"], "gold": {"COUNTRY": ["Afghanistan", "Pakistan", "Bangladesh", "Nepal"], "MUSICAL_INSTRUMENT": [], "BAND": [], "MISC": ["Indian", "Bollywood", "Hindustani"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, BAND, MISC\n\nText:\nDue to shared cultural heritage and language, Indian music and Bollywood films are also popular in Afghanistan, Pakistan, Bangladesh, and Nepal, where Hindustani is widely understood.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, BAND, MISC\n\nText:\nDue to shared cultural heritage and language, Indian music and Bollywood films are also popular in Afghanistan, Pakistan, Bangladesh, and Nepal, where Hindustani is widely understood.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_127", "text": "Several countries geographically outside the boundaries of Europe have competed: Israel, Cyprus and Armenia in Western Asia (Cyprus is a member of the Council of Europe and a member state of the European Union), since 1973, 1981 and Eurovision Song Contest 2006 respectively; Australia since Eurovision Song Contest 2015", "requested_types": ["LOCATION", "COUNTRY", "PERSON", "ORGANISATION", "EVENT", "AWARD"], "distractors": ["AWARD", "PERSON"], "gold": {"LOCATION": ["Europe", "Western Asia"], "COUNTRY": ["Israel", "Cyprus", "Armenia", "Australia"], "PERSON": [], "ORGANISATION": ["Council of Europe", "European Union"], "EVENT": ["Eurovision Song Contest 2006", "Eurovision Song Contest 2015"], "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: LOCATION, COUNTRY, PERSON, ORGANISATION, EVENT, AWARD\n\nText:\nSeveral countries geographically outside the boundaries of Europe have competed: Israel, Cyprus and Armenia in Western Asia (Cyprus is a member of the Council of Europe and a member state of the European Union), since 1973, 1981 and Eurovision Song Contest 2006 respectively; Australia since Eurovision Song Contest 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: LOCATION, COUNTRY, PERSON, ORGANISATION, EVENT, AWARD\n\nText:\nSeveral countries geographically outside the boundaries of Europe have competed: Israel, Cyprus and Armenia in Western Asia (Cyprus is a member of the Council of Europe and a member state of the European Union), since 1973, 1981 and Eurovision Song Contest 2006 respectively; Australia since Eurovision Song Contest 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_music_128", "text": "The 2017 production was nominated for 10 Laurence Olivier Award s and won 2 for Laurence Olivier Award for Best Musical Revival and Laurence Olivier Award for Best Costume Design (by Vicki Mortimer).", "requested_types": ["PERSON", "LOCATION", "COUNTRY", "AWARD"], "distractors": ["LOCATION", "COUNTRY"], "gold": {"PERSON": ["Vicki Mortimer"], "LOCATION": [], "COUNTRY": [], "AWARD": ["Laurence Olivier Award", "Laurence Olivier Award for Best Musical Revival", "Laurence Olivier Award for Best Costume Design"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, AWARD\n\nText:\nThe 2017 production was nominated for 10 Laurence Olivier Award s and won 2 for Laurence Olivier Award for Best Musical Revival and Laurence Olivier Award for Best Costume Design (by Vicki Mortimer).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, AWARD\n\nText:\nThe 2017 production was nominated for 10 Laurence Olivier Award s and won 2 for Laurence Olivier Award for Best Musical Revival and Laurence Olivier Award for Best Costume Design (by Vicki Mortimer).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_129", "text": "While they did not release any studio albums during this period, Devo sporadically reconvened to record a number of songs for various films and compilations, including a cover of the Nine Inch Nails hit Head Like a Hole for the 1992 film Police Story 3: Super Cop and a new recording of Girl U Want on the soundtrack to the 1995 film Tank Girl.", "requested_types": ["SONG", "LOCATION", "ALBUM", "MISC", "BAND"], "distractors": ["ALBUM", "LOCATION"], "gold": {"SONG": ["Head Like a Hole", "Girl U Want"], "LOCATION": [], "ALBUM": [], "MISC": ["Police Story 3: Super Cop", "Tank Girl"], "BAND": ["Devo", "Nine Inch Nails"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, LOCATION, ALBUM, MISC, BAND\n\nText:\nWhile they did not release any studio albums during this period, Devo sporadically reconvened to record a number of songs for various films and compilations, including a cover of the Nine Inch Nails hit Head Like a Hole for the 1992 film Police Story 3: Super Cop and a new recording of Girl U Want on the soundtrack to the 1995 film Tank Girl.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, LOCATION, ALBUM, MISC, BAND\n\nText:\nWhile they did not release any studio albums during this period, Devo sporadically reconvened to record a number of songs for various films and compilations, including a cover of the Nine Inch Nails hit Head Like a Hole for the 1992 film Police Story 3: Super Cop and a new recording of Girl U Want on the soundtrack to the 1995 film Tank Girl.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_130", "text": "The tour opened at the Palace Theatre, Manchester and also played in the Birmingham Hippodrome, the Mayflower Theatre in Southampton, the Edinburgh Playhouse, the Bristol Hippodrome and The Point Theatre in Dublin.", "requested_types": ["EVENT", "BAND", "LOCATION"], "distractors": ["BAND", "EVENT"], "gold": {"EVENT": [], "BAND": [], "LOCATION": ["Palace Theatre, Manchester", "Birmingham Hippodrome", "Mayflower Theatre", "Southampton", "Edinburgh Playhouse", "Bristol Hippodrome", "The Point Theatre", "Dublin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, LOCATION\n\nText:\nThe tour opened at the Palace Theatre, Manchester and also played in the Birmingham Hippodrome, the Mayflower Theatre in Southampton, the Edinburgh Playhouse, the Bristol Hippodrome and The Point Theatre in Dublin.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, LOCATION\n\nText:\nThe tour opened at the Palace Theatre, Manchester and also played in the Birmingham Hippodrome, the Mayflower Theatre in Southampton, the Edinburgh Playhouse, the Bristol Hippodrome and The Point Theatre in Dublin.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_131", "text": "Artists who typified this sound included Travis Tritt, Reba McEntire, George Strait, Keith Whitley, Alan Jackson, Ricky Skaggs, Patty Loveless, Kathy Mattea, Randy Travis, Dwight Yoakam, and The Judds.", "requested_types": ["BAND", "MUSICAL_ARTIST", "ORGANISATION"], "distractors": ["ORGANISATION"], "gold": {"BAND": ["The Judds"], "MUSICAL_ARTIST": ["Travis Tritt", "Reba McEntire", "George Strait", "Keith Whitley", "Alan Jackson", "Ricky Skaggs", "Patty Loveless", "Kathy Mattea", "Randy Travis", "Dwight Yoakam"], "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: BAND, MUSICAL_ARTIST, ORGANISATION\n\nText:\nArtists who typified this sound included Travis Tritt, Reba McEntire, George Strait, Keith Whitley, Alan Jackson, Ricky Skaggs, Patty Loveless, Kathy Mattea, Randy Travis, Dwight Yoakam, and The Judds.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSICAL_ARTIST, ORGANISATION\n\nText:\nArtists who typified this sound included Travis Tritt, Reba McEntire, George Strait, Keith Whitley, Alan Jackson, Ricky Skaggs, Patty Loveless, Kathy Mattea, Randy Travis, Dwight Yoakam, and The Judds.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_132", "text": "However, several 1960s pop music hits feature the bassoon, including The Tears of a Clown by The Miracles (the bassoonist was Charles R. Sirard), Jennifer Juniper by Donovan, 59th Street Bridge Song by Harpers Bizarre, and the oompah bassoon underlying The New Vaudeville Band 's Winchester Cathedral.", "requested_types": ["ALBUM", "MUSICAL_ARTIST", "BAND", "LOCATION", "MISC", "MUSIC_GENRE", "SONG", "PERSON", "MUSICAL_INSTRUMENT"], "distractors": ["ALBUM", "PERSON"], "gold": {"ALBUM": [], "MUSICAL_ARTIST": ["Charles R. Sirard", "Donovan"], "BAND": ["The Miracles", "Harpers Bizarre", "The New Vaudeville Band"], "LOCATION": ["Winchester Cathedral"], "MISC": ["pop music hits"], "MUSIC_GENRE": ["oompah"], "SONG": ["The Tears of a Clown", "Jennifer Juniper", "59th Street Bridge Song"], "PERSON": [], "MUSICAL_INSTRUMENT": ["bassoon"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_ARTIST, BAND, LOCATION, MISC, MUSIC_GENRE, SONG, PERSON, MUSICAL_INSTRUMENT\n\nText:\nHowever, several 1960s pop music hits feature the bassoon, including The Tears of a Clown by The Miracles (the bassoonist was Charles R. Sirard), Jennifer Juniper by Donovan, 59th Street Bridge Song by Harpers Bizarre, and the oompah bassoon underlying The New Vaudeville Band 's Winchester Cathedral.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_ARTIST, BAND, LOCATION, MISC, MUSIC_GENRE, SONG, PERSON, MUSICAL_INSTRUMENT\n\nText:\nHowever, several 1960s pop music hits feature the bassoon, including The Tears of a Clown by The Miracles (the bassoonist was Charles R. Sirard), Jennifer Juniper by Donovan, 59th Street Bridge Song by Harpers Bizarre, and the oompah bassoon underlying The New Vaudeville Band 's Winchester Cathedral.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_133", "text": "The group have released six studio albums: both The Official Fiction (2003) and Desert Lights (2006) topped the ARIA Albums Chart; while Beautiful Sharks (1999), Echolalia (2001) and Leave Your Soul to Science (2012) reached the top 10.", "requested_types": ["PERSON", "MISC", "MUSIC_GENRE", "ALBUM"], "distractors": ["PERSON", "MUSIC_GENRE"], "gold": {"PERSON": [], "MISC": ["ARIA Albums Chart"], "MUSIC_GENRE": [], "ALBUM": ["The Official Fiction", "Desert Lights", "Beautiful Sharks", "Echolalia", "Leave Your Soul to 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, MISC, MUSIC_GENRE, ALBUM\n\nText:\nThe group have released six studio albums: both The Official Fiction (2003) and Desert Lights (2006) topped the ARIA Albums Chart; while Beautiful Sharks (1999), Echolalia (2001) and Leave Your Soul to Science (2012) reached the top 10.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, ALBUM\n\nText:\nThe group have released six studio albums: both The Official Fiction (2003) and Desert Lights (2006) topped the ARIA Albums Chart; while Beautiful Sharks (1999), Echolalia (2001) and Leave Your Soul to Science (2012) reached the top 10.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_134", "text": "Their fifth album, Rosenrot, was released in 2005, and the lead single, Benzin, reached No. 1 in Finland, becoming their first ever No. 1 single.", "requested_types": ["SONG", "MUSICAL_INSTRUMENT", "COUNTRY", "ALBUM", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_INSTRUMENT", "MUSICAL_ARTIST"], "gold": {"SONG": ["Benzin"], "MUSICAL_INSTRUMENT": [], "COUNTRY": ["Finland"], "ALBUM": ["Rosenrot"], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_INSTRUMENT, COUNTRY, ALBUM, MUSICAL_ARTIST\n\nText:\nTheir fifth album, Rosenrot, was released in 2005, and the lead single, Benzin, reached No. 1 in Finland, becoming their first ever No. 1 single.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_INSTRUMENT, COUNTRY, ALBUM, MUSICAL_ARTIST\n\nText:\nTheir fifth album, Rosenrot, was released in 2005, and the lead single, Benzin, reached No. 1 in Finland, becoming their first ever No. 1 single.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_135", "text": "Shoegazing, Stoner rock and noise pop genres emerged into the mainstream with the explosion of bands such as Kyuss, Monster Magnet, the Desert Sessions, Slowdive, the Verve, My Bloody Valentine, Flying Saucer Attack, Loop, Ride, Shiner, the Flaming Lips, Failure, Year of the Rabbit, Cave In, Sun Dial, Hum, Orange Goblin, Porcupine Tree, Spacemen 3, Spiritualized, and Mercury Rev.", "requested_types": ["PERSON", "MUSIC_GENRE", "BAND"], "distractors": ["PERSON"], "gold": {"PERSON": [], "MUSIC_GENRE": ["Shoegazing", "Stoner rock", "noise pop"], "BAND": ["Kyuss", "Monster Magnet", "the Desert Sessions", "Slowdive", "the Verve", "My Bloody Valentine", "Flying Saucer Attack", "Loop", "Ride", "Shiner", "the Flaming Lips", "Failure", "Year of the Rabbit", "Cave In", "Sun Dial", "Hum", "Orange Goblin", "Porcupine Tree", "Spacemen 3", "Spiritualized", "Mercury Rev"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, BAND\n\nText:\nShoegazing, Stoner rock and noise pop genres emerged into the mainstream with the explosion of bands such as Kyuss, Monster Magnet, the Desert Sessions, Slowdive, the Verve, My Bloody Valentine, Flying Saucer Attack, Loop, Ride, Shiner, the Flaming Lips, Failure, Year of the Rabbit, Cave In, Sun Dial, Hum, Orange Goblin, Porcupine Tree, Spacemen 3, Spiritualized, and Mercury Rev.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, BAND\n\nText:\nShoegazing, Stoner rock and noise pop genres emerged into the mainstream with the explosion of bands such as Kyuss, Monster Magnet, the Desert Sessions, Slowdive, the Verve, My Bloody Valentine, Flying Saucer Attack, Loop, Ride, Shiner, the Flaming Lips, Failure, Year of the Rabbit, Cave In, Sun Dial, Hum, Orange Goblin, Porcupine Tree, Spacemen 3, Spiritualized, and Mercury Rev.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_136", "text": "| Invisible Children Invisible Children, Inc, To Write Love On Her Arms, Shirts for a Cure, Keep A Breast Foundation, and Hope For The Day to advocate about their cause.", "requested_types": ["EVENT", "SONG", "MISC", "ORGANISATION"], "distractors": ["EVENT", "SONG", "MISC"], "gold": {"EVENT": [], "SONG": [], "MISC": [], "ORGANISATION": ["Invisible Children, Inc", "To Write Love On Her Arms", "Shirts for a Cure", "Keep A Breast Foundation", "Hope For The Day"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MISC, ORGANISATION\n\nText:\n| Invisible Children Invisible Children, Inc, To Write Love On Her Arms, Shirts for a Cure, Keep A Breast Foundation, and Hope For The Day to advocate about their cause.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MISC, ORGANISATION\n\nText:\n| Invisible Children Invisible Children, Inc, To Write Love On Her Arms, Shirts for a Cure, Keep A Breast Foundation, and Hope For The Day to advocate about their cause.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_137", "text": "House of Pain abruptly broke up in 1996 after the release of their third album, Truth Crushed to Earth Shall Rise Again, which featured guest appearances by rappers Sadat X of Brand Nubian, Guru of Gang Starr, Divine Styler and Cokni O'Dire of the Scheme Team.", "requested_types": ["BAND", "MUSICAL_INSTRUMENT", "MUSIC_GENRE", "MUSICAL_ARTIST", "SONG", "ALBUM"], "distractors": ["SONG", "MUSIC_GENRE", "MUSICAL_INSTRUMENT"], "gold": {"BAND": ["House of Pain", "Brand Nubian", "Gang Starr", "Scheme Team"], "MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": [], "MUSICAL_ARTIST": ["Sadat X", "Guru", "Divine Styler", "Cokni O'Dire"], "SONG": [], "ALBUM": ["Truth Crushed to Earth Shall Rise Again"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSICAL_INSTRUMENT, MUSIC_GENRE, MUSICAL_ARTIST, SONG, ALBUM\n\nText:\nHouse of Pain abruptly broke up in 1996 after the release of their third album, Truth Crushed to Earth Shall Rise Again, which featured guest appearances by rappers Sadat X of Brand Nubian, Guru of Gang Starr, Divine Styler and Cokni O'Dire of the Scheme Team.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSICAL_INSTRUMENT, MUSIC_GENRE, MUSICAL_ARTIST, SONG, ALBUM\n\nText:\nHouse of Pain abruptly broke up in 1996 after the release of their third album, Truth Crushed to Earth Shall Rise Again, which featured guest appearances by rappers Sadat X of Brand Nubian, Guru of Gang Starr, Divine Styler and Cokni O'Dire of the Scheme Team.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_138", "text": "Kid A received a Grammy Award for Grammy Award for Best Alternative Music Album and a nomination for Grammy Award for Album of the Year in early 2001.", "requested_types": ["PERSON", "AWARD", "LOCATION", "ALBUM", "EVENT"], "distractors": ["EVENT", "PERSON", "LOCATION"], "gold": {"PERSON": [], "AWARD": ["Grammy Award", "Grammy Award for Best Alternative Music Album", "Grammy Award for Album of the Year"], "LOCATION": [], "ALBUM": ["Kid A"], "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, AWARD, LOCATION, ALBUM, EVENT\n\nText:\nKid A received a Grammy Award for Grammy Award for Best Alternative Music Album and a nomination for Grammy Award for Album of the Year in early 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: PERSON, AWARD, LOCATION, ALBUM, EVENT\n\nText:\nKid A received a Grammy Award for Grammy Award for Best Alternative Music Album and a nomination for Grammy Award for Album of the Year in early 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_music_139", "text": "Alan Lomax 's recordings of Mississippi Fred McDowell would eventually bring him wider attention on both the blues and Folk music circuit, with McDowell's droning style influencing North Mississippi hill country blues musicians.", "requested_types": ["LOCATION", "MUSIC_GENRE", "MUSICAL_ARTIST", "AWARD"], "distractors": ["AWARD"], "gold": {"LOCATION": ["North Mississippi"], "MUSIC_GENRE": ["blues", "Folk music", "droning style", "hill country blues"], "MUSICAL_ARTIST": ["Alan Lomax", "Mississippi Fred McDowell", "McDowell"], "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: LOCATION, MUSIC_GENRE, MUSICAL_ARTIST, AWARD\n\nText:\nAlan Lomax 's recordings of Mississippi Fred McDowell would eventually bring him wider attention on both the blues and Folk music circuit, with McDowell's droning style influencing North Mississippi hill country blues musicians.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, MUSICAL_ARTIST, AWARD\n\nText:\nAlan Lomax 's recordings of Mississippi Fred McDowell would eventually bring him wider attention on both the blues and Folk music circuit, with McDowell's droning style influencing North Mississippi hill country blues musicians.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_140", "text": "The group has won an Ivor Novello Awards for Outstanding Song Collection from the British Academy of Songwriters, Composers, and Authors, as well as one Grammy Award, two MTV Video Music Award s, and two Billboard Music Award s.", "requested_types": ["ORGANISATION", "MUSICAL_ARTIST", "PERSON", "AWARD"], "distractors": ["PERSON", "MUSICAL_ARTIST"], "gold": {"ORGANISATION": ["British Academy of Songwriters, Composers, and Authors"], "MUSICAL_ARTIST": [], "PERSON": [], "AWARD": ["Ivor Novello Awards for Outstanding Song Collection", "Grammy Award", "MTV Video Music Award", "Billboard Music 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, MUSICAL_ARTIST, PERSON, AWARD\n\nText:\nThe group has won an Ivor Novello Awards for Outstanding Song Collection from the British Academy of Songwriters, Composers, and Authors, as well as one Grammy Award, two MTV Video Music Award s, and two Billboard Music Award 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, MUSICAL_ARTIST, PERSON, AWARD\n\nText:\nThe group has won an Ivor Novello Awards for Outstanding Song Collection from the British Academy of Songwriters, Composers, and Authors, as well as one Grammy Award, two MTV Video Music Award s, and two Billboard Music Award 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_music_141", "text": "In 1998 Warp signed Boards Of Canada, a duo that would go on to release some of the most highly revered electronic music albums of their time: Music Has the Right to Children (1998), Geogaddi (2002), The Campfire Headphase (2005) and Tomorrow's Harvest (2013).", "requested_types": ["ALBUM", "LOCATION", "BAND", "MUSIC_GENRE", "MUSICAL_ARTIST"], "distractors": ["LOCATION"], "gold": {"ALBUM": ["Music Has the Right to Children", "Geogaddi", "The Campfire Headphase", "Tomorrow's Harvest"], "LOCATION": [], "BAND": ["Boards Of Canada"], "MUSIC_GENRE": ["electronic music"], "MUSICAL_ARTIST": ["Warp"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, LOCATION, BAND, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nIn 1998 Warp signed Boards Of Canada, a duo that would go on to release some of the most highly revered electronic music albums of their time: Music Has the Right to Children (1998), Geogaddi (2002), The Campfire Headphase (2005) and Tomorrow's Harvest (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: ALBUM, LOCATION, BAND, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nIn 1998 Warp signed Boards Of Canada, a duo that would go on to release some of the most highly revered electronic music albums of their time: Music Has the Right to Children (1998), Geogaddi (2002), The Campfire Headphase (2005) and Tomorrow's Harvest (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_music_142", "text": "Three of them, Emergency on Planet Earth (1993), Synkronized (1999), A Funk Odyssey (2001), along with their greatest hits compilation, charted at number 1.", "requested_types": ["MUSICAL_INSTRUMENT", "COUNTRY", "ALBUM", "EVENT"], "distractors": ["COUNTRY", "MUSICAL_INSTRUMENT", "EVENT"], "gold": {"MUSICAL_INSTRUMENT": [], "COUNTRY": [], "ALBUM": ["Emergency on Planet Earth", "Synkronized", "A Funk Odyssey", "their greatest hits"], "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: MUSICAL_INSTRUMENT, COUNTRY, ALBUM, EVENT\n\nText:\nThree of them, Emergency on Planet Earth (1993), Synkronized (1999), A Funk Odyssey (2001), along with their greatest hits compilation, charted at number 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: MUSICAL_INSTRUMENT, COUNTRY, ALBUM, EVENT\n\nText:\nThree of them, Emergency on Planet Earth (1993), Synkronized (1999), A Funk Odyssey (2001), along with their greatest hits compilation, charted at number 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_music_143", "text": "Link Wray Biography, AllMusic.com; accessed March 17, 2015 facilitating the emergence of Punk rock and Hard rock.", "requested_types": ["MUSIC_GENRE", "COUNTRY", "MUSICAL_ARTIST", "LOCATION"], "distractors": ["COUNTRY", "LOCATION"], "gold": {"MUSIC_GENRE": ["Punk rock", "Hard rock"], "COUNTRY": [], "MUSICAL_ARTIST": ["Link Wray"], "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: MUSIC_GENRE, COUNTRY, MUSICAL_ARTIST, LOCATION\n\nText:\nLink Wray Biography, AllMusic.com; accessed March 17, 2015 facilitating the emergence of Punk rock and Hard rock.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, COUNTRY, MUSICAL_ARTIST, LOCATION\n\nText:\nLink Wray Biography, AllMusic.com; accessed March 17, 2015 facilitating the emergence of Punk rock and Hard rock.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_144", "text": "In the Southwestern United States, it was the Rocky Mountains, American frontier, and Rio Grande that acted as a similar backdrop for Indigenous music of North America, Mexican, and cowboy ballads, which resulted in New Mexico music and the development of Western music, and its directly related Red Dirt, Texas country music, and Tejano music styles.", "requested_types": ["COUNTRY", "LOCATION", "MISC", "MUSIC_GENRE"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "LOCATION": ["Southwestern United States", "Rocky Mountains", "American frontier", "Rio Grande"], "MISC": ["Mexican"], "MUSIC_GENRE": ["Indigenous music of North America", "cowboy ballads", "New Mexico music", "Western music", "Red Dirt", "Texas country music", "Tejano music"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, MUSIC_GENRE\n\nText:\nIn the Southwestern United States, it was the Rocky Mountains, American frontier, and Rio Grande that acted as a similar backdrop for Indigenous music of North America, Mexican, and cowboy ballads, which resulted in New Mexico music and the development of Western music, and its directly related Red Dirt, Texas country music, and Tejano music styles.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, MUSIC_GENRE\n\nText:\nIn the Southwestern United States, it was the Rocky Mountains, American frontier, and Rio Grande that acted as a similar backdrop for Indigenous music of North America, Mexican, and cowboy ballads, which resulted in New Mexico music and the development of Western music, and its directly related Red Dirt, Texas country music, and Tejano music styles.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_145", "text": "He also appeared with RBX, Nas and KRS-One on East Coast Killer, West Coast Killer from Dr. Dre's Dr. Dre Presents the Aftermath album, and contributed to an album entitled The Psycho Realm with Psycho Realm.", "requested_types": ["ALBUM", "SONG", "BAND", "COUNTRY", "MUSICAL_ARTIST"], "distractors": ["COUNTRY"], "gold": {"ALBUM": ["Dr. Dre Presents the Aftermath", "The Psycho Realm"], "SONG": ["East Coast Killer, West Coast Killer"], "BAND": ["Psycho Realm"], "COUNTRY": [], "MUSICAL_ARTIST": ["RBX", "Nas", "KRS-One", "Dr. Dre"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, SONG, BAND, COUNTRY, MUSICAL_ARTIST\n\nText:\nHe also appeared with RBX, Nas and KRS-One on East Coast Killer, West Coast Killer from Dr. Dre's Dr. Dre Presents the Aftermath album, and contributed to an album entitled The Psycho Realm with Psycho Realm.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, SONG, BAND, COUNTRY, MUSICAL_ARTIST\n\nText:\nHe also appeared with RBX, Nas and KRS-One on East Coast Killer, West Coast Killer from Dr. Dre's Dr. Dre Presents the Aftermath album, and contributed to an album entitled The Psycho Realm with Psycho Realm.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_146", "text": "Thumrī is also used as a generic name for some other, even lighter, forms such as Dadra, Hori, Kajari, Sawani, Jhoola, and Chaiti, even though each of them has its own structure and content - either lyrical or musical or both - and so the exposition of these forms vary.", "requested_types": ["MUSIC_GENRE", "BAND"], "distractors": ["BAND"], "gold": {"MUSIC_GENRE": ["Thumrī", "Dadra", "Hori", "Kajari", "Sawani", "Jhoola", "Chaiti"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, BAND\n\nText:\nThumrī is also used as a generic name for some other, even lighter, forms such as Dadra, Hori, Kajari, Sawani, Jhoola, and Chaiti, even though each of them has its own structure and content - either lyrical or musical or both - and so the exposition of these forms vary.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, BAND\n\nText:\nThumrī is also used as a generic name for some other, even lighter, forms such as Dadra, Hori, Kajari, Sawani, Jhoola, and Chaiti, even though each of them has its own structure and content - either lyrical or musical or both - and so the exposition of these forms vary.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_147", "text": "At the time of his sixteenth birthday on June 20, 1958, Brian Wilson shared a bedroom with his brothers, Dennis Wilson and Carl Wilson - aged thirteen and eleven, respectively - in their family home in Hawthorne.", "requested_types": ["MUSICAL_ARTIST", "LOCATION", "ORGANISATION", "BAND"], "distractors": ["BAND", "ORGANISATION"], "gold": {"MUSICAL_ARTIST": ["Brian Wilson", "Dennis Wilson", "Carl Wilson"], "LOCATION": ["Hawthorne"], "ORGANISATION": [], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, LOCATION, ORGANISATION, BAND\n\nText:\nAt the time of his sixteenth birthday on June 20, 1958, Brian Wilson shared a bedroom with his brothers, Dennis Wilson and Carl Wilson - aged thirteen and eleven, respectively - in their family home in Hawthorne.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, LOCATION, ORGANISATION, BAND\n\nText:\nAt the time of his sixteenth birthday on June 20, 1958, Brian Wilson shared a bedroom with his brothers, Dennis Wilson and Carl Wilson - aged thirteen and eleven, respectively - in their family home in Hawthorne.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_148", "text": "Wills favored jazz-like arrangements and the band found national popularity into the 1940s with such hits as Steel Guitar Rag, New San Antonio Rose, Smoke On The Water, Stars And Stripes On Iwo Jima, and New Spanish Two Step.", "requested_types": ["MUSICAL_ARTIST", "SONG", "MUSICAL_INSTRUMENT", "MUSIC_GENRE", "COUNTRY"], "distractors": ["COUNTRY", "MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_ARTIST": ["Wills"], "SONG": ["Steel Guitar Rag", "New San Antonio Rose", "Smoke On The Water", "Stars And Stripes On Iwo Jima", "New Spanish Two Step"], "MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": ["jazz-like"], "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: MUSICAL_ARTIST, SONG, MUSICAL_INSTRUMENT, MUSIC_GENRE, COUNTRY\n\nText:\nWills favored jazz-like arrangements and the band found national popularity into the 1940s with such hits as Steel Guitar Rag, New San Antonio Rose, Smoke On The Water, Stars And Stripes On Iwo Jima, and New Spanish Two Step.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, SONG, MUSICAL_INSTRUMENT, MUSIC_GENRE, COUNTRY\n\nText:\nWills favored jazz-like arrangements and the band found national popularity into the 1940s with such hits as Steel Guitar Rag, New San Antonio Rose, Smoke On The Water, Stars And Stripes On Iwo Jima, and New Spanish Two Step.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_149", "text": "(e.g. Red Sector A, Closer to the Heart on A Show of Hands and Mystic Rhythms on R30.", "requested_types": ["ALBUM", "PERSON", "MUSIC_GENRE", "EVENT", "SONG", "MISC"], "distractors": ["MISC", "MUSIC_GENRE", "PERSON"], "gold": {"ALBUM": ["A Show of Hands"], "PERSON": [], "MUSIC_GENRE": [], "EVENT": ["R30"], "SONG": ["Red Sector A", "Closer to the Heart", "Mystic Rhythms"], "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: ALBUM, PERSON, MUSIC_GENRE, EVENT, SONG, MISC\n\nText:\n(e.g. Red Sector A, Closer to the Heart on A Show of Hands and Mystic Rhythms on R30.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, PERSON, MUSIC_GENRE, EVENT, SONG, MISC\n\nText:\n(e.g. Red Sector A, Closer to the Heart on A Show of Hands and Mystic Rhythms on R30.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_150", "text": "The new millennium brought about a drastic change in Autechre's style, demonstrated by Confield (2001) and Draft 7.30 (2003), as well as the Gantz Graf EP (2002).", "requested_types": ["MISC", "SONG", "BAND", "COUNTRY", "ALBUM", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_ARTIST", "COUNTRY", "SONG"], "gold": {"MISC": ["EP"], "SONG": [], "BAND": ["Autechre"], "COUNTRY": [], "ALBUM": ["Confield", "Draft 7.30", "Gantz Graf"], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, BAND, COUNTRY, ALBUM, MUSICAL_ARTIST\n\nText:\nThe new millennium brought about a drastic change in Autechre's style, demonstrated by Confield (2001) and Draft 7.30 (2003), as well as the Gantz Graf EP (2002).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, BAND, COUNTRY, ALBUM, MUSICAL_ARTIST\n\nText:\nThe new millennium brought about a drastic change in Autechre's style, demonstrated by Confield (2001) and Draft 7.30 (2003), as well as the Gantz Graf EP (2002).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_151", "text": "The agreement sees an annual fixture at the MCG, beginning with a clash between Australia and European champions Greece on 25 May 2006 in front of a sell-out crowd of 95,103, before Australia left to contest in the World Cup finals.", "requested_types": ["EVENT", "COUNTRY", "AWARD", "MISC", "LOCATION"], "distractors": ["AWARD"], "gold": {"EVENT": ["World Cup"], "COUNTRY": ["Australia", "Greece"], "AWARD": [], "MISC": ["European"], "LOCATION": ["MCG"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, AWARD, MISC, LOCATION\n\nText:\nThe agreement sees an annual fixture at the MCG, beginning with a clash between Australia and European champions Greece on 25 May 2006 in front of a sell-out crowd of 95,103, before Australia left to contest in the World Cup finals.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, AWARD, MISC, LOCATION\n\nText:\nThe agreement sees an annual fixture at the MCG, beginning with a clash between Australia and European champions Greece on 25 May 2006 in front of a sell-out crowd of 95,103, before Australia left to contest in the World Cup finals.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_152", "text": "As the recording industry grew, country blues performers like Bo Carter, Jimmie Rodgers (country singer), Blind Lemon Jefferson, Lonnie Johnson, Tampa Red and Blind Blake became more popular in the African American community.", "requested_types": ["EVENT", "MUSIC_GENRE", "LOCATION", "MISC", "AWARD", "MUSICAL_ARTIST"], "distractors": ["AWARD", "LOCATION", "EVENT"], "gold": {"EVENT": [], "MUSIC_GENRE": ["country blues"], "LOCATION": [], "MISC": ["country singer", "African American"], "AWARD": [], "MUSICAL_ARTIST": ["Bo Carter", "Jimmie Rodgers", "Blind Lemon Jefferson", "Lonnie Johnson", "Tampa Red", "Blind Blake"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, LOCATION, MISC, AWARD, MUSICAL_ARTIST\n\nText:\nAs the recording industry grew, country blues performers like Bo Carter, Jimmie Rodgers (country singer), Blind Lemon Jefferson, Lonnie Johnson, Tampa Red and Blind Blake became more popular in the African American community.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, LOCATION, MISC, AWARD, MUSICAL_ARTIST\n\nText:\nAs the recording industry grew, country blues performers like Bo Carter, Jimmie Rodgers (country singer), Blind Lemon Jefferson, Lonnie Johnson, Tampa Red and Blind Blake became more popular in the African American community.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_153", "text": "Some of Mattacks' most notable participation in studio recordings in the late 1970s are the work on art rock studio albums by Brian Eno (Before and After Science) and 801 's Listen Now, as well as several Ashley Hutchings -related folk rock projects (The Compleat Dancing Master, Son of Morris On etc.", "requested_types": ["BAND", "SONG", "ALBUM", "MISC", "EVENT", "MUSIC_GENRE", "ORGANISATION", "MUSICAL_ARTIST"], "distractors": ["SONG", "EVENT", "ORGANISATION"], "gold": {"BAND": ["801"], "SONG": [], "ALBUM": ["Before and After Science", "Listen Now", "The Compleat Dancing Master", "Son of Morris On"], "MISC": ["art", "rock", "studio", "albums"], "EVENT": [], "MUSIC_GENRE": ["folk rock"], "ORGANISATION": [], "MUSICAL_ARTIST": ["Mattacks", "Brian Eno", "Ashley Hutchings"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, SONG, ALBUM, MISC, EVENT, MUSIC_GENRE, ORGANISATION, MUSICAL_ARTIST\n\nText:\nSome of Mattacks' most notable participation in studio recordings in the late 1970s are the work on art rock studio albums by Brian Eno (Before and After Science) and 801 's Listen Now, as well as several Ashley Hutchings -related folk rock projects (The Compleat Dancing Master, Son of Morris On 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: BAND, SONG, ALBUM, MISC, EVENT, MUSIC_GENRE, ORGANISATION, MUSICAL_ARTIST\n\nText:\nSome of Mattacks' most notable participation in studio recordings in the late 1970s are the work on art rock studio albums by Brian Eno (Before and After Science) and 801 's Listen Now, as well as several Ashley Hutchings -related folk rock projects (The Compleat Dancing Master, Son of Morris On 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_music_154", "text": "Contemporary African-American performers who work in this style of the blues include Bobby Rush, Denise LaSalle, Sir Charles Jones, Bettye LaVette, Marvin Sease, Peggy Scott-Adams, Mel Waiters, Clarence Carter, Dr. Feelgood Potts, O.B. Buchana, Ms. Jody, Shirley Brown, and dozens of others.", "requested_types": ["MUSIC_GENRE", "MUSICAL_ARTIST", "MISC", "PERSON", "ALBUM", "COUNTRY"], "distractors": ["PERSON", "COUNTRY", "ALBUM"], "gold": {"MUSIC_GENRE": ["blues"], "MUSICAL_ARTIST": ["Bobby Rush", "Denise LaSalle", "Sir Charles Jones", "Bettye LaVette", "Marvin Sease", "Peggy Scott-Adams", "Mel Waiters", "Clarence Carter", "Dr. Feelgood Potts", "O.B. Buchana", "Ms. Jody", "Shirley Brown"], "MISC": ["African-American"], "PERSON": [], "ALBUM": [], "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: MUSIC_GENRE, MUSICAL_ARTIST, MISC, PERSON, ALBUM, COUNTRY\n\nText:\nContemporary African-American performers who work in this style of the blues include Bobby Rush, Denise LaSalle, Sir Charles Jones, Bettye LaVette, Marvin Sease, Peggy Scott-Adams, Mel Waiters, Clarence Carter, Dr. Feelgood Potts, O.B. Buchana, Ms. Jody, Shirley Brown, and dozens of 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: MUSIC_GENRE, MUSICAL_ARTIST, MISC, PERSON, ALBUM, COUNTRY\n\nText:\nContemporary African-American performers who work in this style of the blues include Bobby Rush, Denise LaSalle, Sir Charles Jones, Bettye LaVette, Marvin Sease, Peggy Scott-Adams, Mel Waiters, Clarence Carter, Dr. Feelgood Potts, O.B. Buchana, Ms. Jody, Shirley Brown, and dozens of 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_music_155", "text": "Wilson's charity work includes the Susan G. Komen for the Cure Race for the Cure, the American Cancer Society, St. Jude's Children's Research Hospital, the Easter Seals Foundation, UNICEF, The NAACP, the Cystic Fibrosis Foundation, the All-Star Network, and Figure Skaters of Harlem, a youth organization devoted to helping children towards entering the Olympics.", "requested_types": ["LOCATION", "ORGANISATION", "EVENT", "MUSICAL_ARTIST"], "distractors": ["LOCATION"], "gold": {"LOCATION": [], "ORGANISATION": ["Susan G. Komen for the Cure", "American Cancer Society", "St. Jude's Children's Research Hospital", "Easter Seals Foundation", "UNICEF", "NAACP", "Cystic Fibrosis Foundation", "All-Star Network", "Figure Skaters of Harlem"], "EVENT": ["Race for the Cure", "Olympics"], "MUSICAL_ARTIST": ["Wilson"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST\n\nText:\nWilson's charity work includes the Susan G. Komen for the Cure Race for the Cure, the American Cancer Society, St. Jude's Children's Research Hospital, the Easter Seals Foundation, UNICEF, The NAACP, the Cystic Fibrosis Foundation, the All-Star Network, and Figure Skaters of Harlem, a youth organization devoted to helping children towards entering the 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, MUSICAL_ARTIST\n\nText:\nWilson's charity work includes the Susan G. Komen for the Cure Race for the Cure, the American Cancer Society, St. Jude's Children's Research Hospital, the Easter Seals Foundation, UNICEF, The NAACP, the Cystic Fibrosis Foundation, the All-Star Network, and Figure Skaters of Harlem, a youth organization devoted to helping children towards entering the 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_music_156", "text": "In 1990, he appeared on Kool Thing, a song by the alternative rock band Sonic Youth, and along with Flavor Flav, he sang on George Clinton 's song Tweakin', which appears on his 1989 album The Cinderella Theory.", "requested_types": ["MUSICAL_ARTIST", "ALBUM", "LOCATION", "BAND", "SONG", "MUSIC_GENRE"], "distractors": ["LOCATION"], "gold": {"MUSICAL_ARTIST": ["Flavor Flav", "George Clinton"], "ALBUM": ["The Cinderella Theory"], "LOCATION": [], "BAND": ["Sonic Youth"], "SONG": ["Kool Thing", "Tweakin"], "MUSIC_GENRE": ["alternative rock"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ALBUM, LOCATION, BAND, SONG, MUSIC_GENRE\n\nText:\nIn 1990, he appeared on Kool Thing, a song by the alternative rock band Sonic Youth, and along with Flavor Flav, he sang on George Clinton 's song Tweakin', which appears on his 1989 album The Cinderella Theory.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ALBUM, LOCATION, BAND, SONG, MUSIC_GENRE\n\nText:\nIn 1990, he appeared on Kool Thing, a song by the alternative rock band Sonic Youth, and along with Flavor Flav, he sang on George Clinton 's song Tweakin', which appears on his 1989 album The Cinderella Theory.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_157", "text": "After Deftones' third album White Pony, subsequent releases would be written with seven-strings, until 2010's Diamond Eyes and 2012's Koi No Yokan, which were written with an eight-string guitar.", "requested_types": ["PERSON", "AWARD", "BAND", "ALBUM", "MUSICAL_INSTRUMENT"], "distractors": ["AWARD", "PERSON"], "gold": {"PERSON": [], "AWARD": [], "BAND": ["Deftones"], "ALBUM": ["White Pony", "Diamond Eyes", "Koi No Yokan"], "MUSICAL_INSTRUMENT": ["seven-strings", "eight-string guitar"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM, MUSICAL_INSTRUMENT\n\nText:\nAfter Deftones' third album White Pony, subsequent releases would be written with seven-strings, until 2010's Diamond Eyes and 2012's Koi No Yokan, which were written with an eight-string guitar.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM, MUSICAL_INSTRUMENT\n\nText:\nAfter Deftones' third album White Pony, subsequent releases would be written with seven-strings, until 2010's Diamond Eyes and 2012's Koi No Yokan, which were written with an eight-string guitar.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_158", "text": "The four-piece Indie rock band played gigs at pubs and festivals from 2005 to 2007 such as Knitting Factory, Bamboozle Left, The Roxy, Spaceland, and The Viper Room.", "requested_types": ["LOCATION", "MISC", "MUSICAL_ARTIST", "MUSIC_GENRE"], "distractors": ["MISC", "MUSICAL_ARTIST"], "gold": {"LOCATION": ["Knitting Factory", "Bamboozle Left", "The Roxy", "Spaceland", "The Viper Room"], "MISC": [], "MUSICAL_ARTIST": [], "MUSIC_GENRE": ["Indie rock"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nThe four-piece Indie rock band played gigs at pubs and festivals from 2005 to 2007 such as Knitting Factory, Bamboozle Left, The Roxy, Spaceland, and The Viper Room.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nThe four-piece Indie rock band played gigs at pubs and festivals from 2005 to 2007 such as Knitting Factory, Bamboozle Left, The Roxy, Spaceland, and The Viper Room.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_159", "text": "His mix of traditional Senegalese mbalax with eclectic influences ranging from Cuban rumba to Hip hop music, jazz, and Soul music won him an international fan base of millions.", "requested_types": ["MISC", "MUSIC_GENRE", "BAND"], "distractors": ["BAND"], "gold": {"MISC": ["Senegalese"], "MUSIC_GENRE": ["mbalax", "eclectic", "Cuban rumba", "Hip hop music", "jazz", "Soul music"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, BAND\n\nText:\nHis mix of traditional Senegalese mbalax with eclectic influences ranging from Cuban rumba to Hip hop music, jazz, and Soul music won him an international fan base of millions.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, BAND\n\nText:\nHis mix of traditional Senegalese mbalax with eclectic influences ranging from Cuban rumba to Hip hop music, jazz, and Soul music won him an international fan base of millions.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_160", "text": "This list includes artists that perform in traditional gospel music genres such as Southern gospel, traditional black gospel, progressive Southern gospel, urban contemporary gospel, gospel blues, Christian country music, Celtic gospel and British black gospel as well as artists in the general market who have recorded music in these genres.", "requested_types": ["COUNTRY", "MUSIC_GENRE"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "MUSIC_GENRE": ["gospel music", "Southern gospel", "traditional black gospel", "progressive Southern gospel", "urban contemporary gospel", "gospel blues", "Christian country music", "Celtic gospel", "British black gospel"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE\n\nText:\nThis list includes artists that perform in traditional gospel music genres such as Southern gospel, traditional black gospel, progressive Southern gospel, urban contemporary gospel, gospel blues, Christian country music, Celtic gospel and British black gospel as well as artists in the general market who have recorded music in these genres.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE\n\nText:\nThis list includes artists that perform in traditional gospel music genres such as Southern gospel, traditional black gospel, progressive Southern gospel, urban contemporary gospel, gospel blues, Christian country music, Celtic gospel and British black gospel as well as artists in the general market who have recorded music in these genres.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_161", "text": "Dave Bush joined on keyboards for 1992's Code: Selfish, followed by the band's return to an independent record label for The Infotainment Scan (1993), Middle Class Revolt (1994), and Cerebral Caustic (1995).", "requested_types": ["MISC", "ALBUM", "MUSICAL_ARTIST", "LOCATION"], "distractors": ["MISC", "LOCATION"], "gold": {"MISC": [], "ALBUM": ["Code: Selfish", "The Infotainment Scan", "Middle Class Revolt", "Cerebral Caustic"], "MUSICAL_ARTIST": ["Dave Bush"], "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, ALBUM, MUSICAL_ARTIST, LOCATION\n\nText:\nDave Bush joined on keyboards for 1992's Code: Selfish, followed by the band's return to an independent record label for The Infotainment Scan (1993), Middle Class Revolt (1994), and Cerebral Caustic (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: MISC, ALBUM, MUSICAL_ARTIST, LOCATION\n\nText:\nDave Bush joined on keyboards for 1992's Code: Selfish, followed by the band's return to an independent record label for The Infotainment Scan (1993), Middle Class Revolt (1994), and Cerebral Caustic (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_music_162", "text": "The Nordic Choir has also appeared throughout the United States, performing in well-known concert halls as Lincoln Center in New York and the John F. Kennedy Center for the Performing Arts in Washington, D.C. Additionally, Luther College has the largest collegiate choral program in the United States with almost 600 student singers.", "requested_types": ["COUNTRY", "PERSON", "BAND", "LOCATION"], "distractors": ["PERSON"], "gold": {"COUNTRY": ["United States"], "PERSON": [], "BAND": ["Nordic Choir"], "LOCATION": ["Lincoln Center", "New York", "John F. Kennedy Center for the Performing Arts", "Washington, D.C.", "Luther College"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, BAND, LOCATION\n\nText:\nThe Nordic Choir has also appeared throughout the United States, performing in well-known concert halls as Lincoln Center in New York and the John F. Kennedy Center for the Performing Arts in Washington, D.C. Additionally, Luther College has the largest collegiate choral program in the United States with almost 600 student 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: COUNTRY, PERSON, BAND, LOCATION\n\nText:\nThe Nordic Choir has also appeared throughout the United States, performing in well-known concert halls as Lincoln Center in New York and the John F. Kennedy Center for the Performing Arts in Washington, D.C. Additionally, Luther College has the largest collegiate choral program in the United States with almost 600 student 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_music_163", "text": "Baron Cohen was named Best Newcomer at the 1999 British Comedy Awards for The 11 O'Clock Show, and since then, he has received two British Academy of Film and Television Arts Awards for Da Ali G Show, several Emmy nominations, a nomination for an Academy Award for Best Adapted Screenplay, and a Golden Globe for Best Actor for his work in the feature film Borat.", "requested_types": ["EVENT", "MUSIC_GENRE", "MUSICAL_INSTRUMENT", "PERSON", "MISC", "AWARD"], "distractors": ["MUSICAL_INSTRUMENT", "MUSIC_GENRE"], "gold": {"EVENT": ["The 11 O'Clock Show", "Da Ali G Show"], "MUSIC_GENRE": [], "MUSICAL_INSTRUMENT": [], "PERSON": ["Baron Cohen"], "MISC": ["Borat"], "AWARD": ["Best Newcomer at the 1999 British Comedy Awards", "British Academy of Film and Television Arts Awards", "Emmy", "Academy Award for Best Adapted Screenplay", "Golden Globe for Best Actor"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, MUSICAL_INSTRUMENT, PERSON, MISC, AWARD\n\nText:\nBaron Cohen was named Best Newcomer at the 1999 British Comedy Awards for The 11 O'Clock Show, and since then, he has received two British Academy of Film and Television Arts Awards for Da Ali G Show, several Emmy nominations, a nomination for an Academy Award for Best Adapted Screenplay, and a Golden Globe for Best Actor for his work in the feature film Borat.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, MUSICAL_INSTRUMENT, PERSON, MISC, AWARD\n\nText:\nBaron Cohen was named Best Newcomer at the 1999 British Comedy Awards for The 11 O'Clock Show, and since then, he has received two British Academy of Film and Television Arts Awards for Da Ali G Show, several Emmy nominations, a nomination for an Academy Award for Best Adapted Screenplay, and a Golden Globe for Best Actor for his work in the feature film Borat.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_164", "text": "Australian Film Institute, Blockbuster Entertainment Awards, Empire Awards, Hollywood Film Festival, London Film Critics' Circle, Russian Guild of Film Critics, Satellite Awards, and Southeastern Film Critics Association.", "requested_types": ["AWARD", "ORGANISATION", "COUNTRY", "EVENT"], "distractors": ["COUNTRY"], "gold": {"AWARD": ["Blockbuster Entertainment Awards", "Empire Awards", "Satellite Awards"], "ORGANISATION": ["Australian Film Institute", "London Film Critics' Circle", "Russian Guild of Film Critics", "Southeastern Film Critics Association"], "COUNTRY": [], "EVENT": ["Hollywood Film Festival"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, EVENT\n\nText:\nAustralian Film Institute, Blockbuster Entertainment Awards, Empire Awards, Hollywood Film Festival, London Film Critics' Circle, Russian Guild of Film Critics, Satellite Awards, and Southeastern Film Critics 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: AWARD, ORGANISATION, COUNTRY, EVENT\n\nText:\nAustralian Film Institute, Blockbuster Entertainment Awards, Empire Awards, Hollywood Film Festival, London Film Critics' Circle, Russian Guild of Film Critics, Satellite Awards, and Southeastern Film Critics 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_music_165", "text": "This was the instrument of the early jazz great Johnny St. Cyr, jazzmen Django Reinhardt, Danny Barker, Papa Charlie Jackson and Clancy Hayes, as well as the blues and gospel singer The Reverend Gary Davis.", "requested_types": ["ORGANISATION", "MUSIC_GENRE", "MUSICAL_ARTIST", "SONG", "MISC"], "distractors": ["SONG", "MISC", "ORGANISATION"], "gold": {"ORGANISATION": [], "MUSIC_GENRE": ["jazz", "blues", "gospel"], "MUSICAL_ARTIST": ["Johnny St. Cyr", "Django Reinhardt", "Danny Barker", "Papa Charlie Jackson", "Clancy Hayes", "Reverend Gary Davis"], "SONG": [], "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, MUSIC_GENRE, MUSICAL_ARTIST, SONG, MISC\n\nText:\nThis was the instrument of the early jazz great Johnny St. Cyr, jazzmen Django Reinhardt, Danny Barker, Papa Charlie Jackson and Clancy Hayes, as well as the blues and gospel singer The Reverend Gary Davis.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, MUSICAL_ARTIST, SONG, MISC\n\nText:\nThis was the instrument of the early jazz great Johnny St. Cyr, jazzmen Django Reinhardt, Danny Barker, Papa Charlie Jackson and Clancy Hayes, as well as the blues and gospel singer The Reverend Gary Davis.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_166", "text": "For portraying both gunfighter Kid Shelleen and criminal Tim Strawn, he won the Academy Award for Best Actor, along with a BAFTA Award for Best Actor in a Leading Role, a Golden Globe Award, an National Board of Review Award for Best Actor, and the Silver Bear for Best Actor.", "requested_types": ["AWARD", "BAND", "PERSON"], "distractors": ["BAND"], "gold": {"AWARD": ["Academy Award for Best Actor", "BAFTA Award for Best Actor in a Leading Role", "Golden Globe Award", "National Board of Review Award for Best Actor", "Silver Bear for Best Actor"], "BAND": [], "PERSON": ["Kid Shelleen", "Tim Strawn"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, PERSON\n\nText:\nFor portraying both gunfighter Kid Shelleen and criminal Tim Strawn, he won the Academy Award for Best Actor, along with a BAFTA Award for Best Actor in a Leading Role, a Golden Globe Award, an National Board of Review Award for Best Actor, and the Silver Bear for Best Actor.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, PERSON\n\nText:\nFor portraying both gunfighter Kid Shelleen and criminal Tim Strawn, he won the Academy Award for Best Actor, along with a BAFTA Award for Best Actor in a Leading Role, a Golden Globe Award, an National Board of Review Award for Best Actor, and the Silver Bear for Best Actor.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_167", "text": "This resulted in the recordings of successful and critically acclaimed albums such as Marvin Gaye 's What's Going On (1971) and Let's Get it On (1973), and Stevie Wonder' s Music of My Mind (1972), Talking Book (1972), and Innervisions (1973).", "requested_types": ["LOCATION", "ALBUM", "MISC", "EVENT", "MUSICAL_ARTIST"], "distractors": ["MISC", "EVENT", "LOCATION"], "gold": {"LOCATION": [], "ALBUM": ["What's Going On", "Let's Get it On", "Music of My Mind", "Talking Book", "Innervisions"], "MISC": [], "EVENT": [], "MUSICAL_ARTIST": ["Marvin Gaye", "Stevie Wonder"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MISC, EVENT, MUSICAL_ARTIST\n\nText:\nThis resulted in the recordings of successful and critically acclaimed albums such as Marvin Gaye 's What's Going On (1971) and Let's Get it On (1973), and Stevie Wonder' s Music of My Mind (1972), Talking Book (1972), and Innervisions (1973).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MISC, EVENT, MUSICAL_ARTIST\n\nText:\nThis resulted in the recordings of successful and critically acclaimed albums such as Marvin Gaye 's What's Going On (1971) and Let's Get it On (1973), and Stevie Wonder' s Music of My Mind (1972), Talking Book (1972), and Innervisions (1973).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_168", "text": "Styles of music unique to the nation of Haiti include music derived from Vodou ceremonial traditions, Rara parading music, Twoubadou ballads, Mini-jazz rock bands, Rasin movement, Hip hop Kreyòl, the wildly popular Compas, and Méringue as its basic rhythm.", "requested_types": ["ALBUM", "MISC", "COUNTRY", "MUSIC_GENRE"], "distractors": ["ALBUM"], "gold": {"ALBUM": [], "MISC": ["Vodou", "ceremonial", "traditions"], "COUNTRY": ["Haiti"], "MUSIC_GENRE": ["Rara parading music", "Twoubadou ballads", "Mini-jazz rock", "Rasin", "Hip hop Kreyòl", "Compas", "Méringue"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MISC, COUNTRY, MUSIC_GENRE\n\nText:\nStyles of music unique to the nation of Haiti include music derived from Vodou ceremonial traditions, Rara parading music, Twoubadou ballads, Mini-jazz rock bands, Rasin movement, Hip hop Kreyòl, the wildly popular Compas, and Méringue as its basic rhythm.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MISC, COUNTRY, MUSIC_GENRE\n\nText:\nStyles of music unique to the nation of Haiti include music derived from Vodou ceremonial traditions, Rara parading music, Twoubadou ballads, Mini-jazz rock bands, Rasin movement, Hip hop Kreyòl, the wildly popular Compas, and Méringue as its basic rhythm.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_169", "text": "Rohani was commissioned in 1998 by the government of Thailand and the committee of the 1998 Asian Games to compose and conduct the music for opening ceremonies.", "requested_types": ["PERSON", "MUSIC_GENRE", "COUNTRY", "EVENT"], "distractors": ["MUSIC_GENRE"], "gold": {"PERSON": ["Rohani"], "MUSIC_GENRE": [], "COUNTRY": ["Thailand"], "EVENT": ["1998 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: PERSON, MUSIC_GENRE, COUNTRY, EVENT\n\nText:\nRohani was commissioned in 1998 by the government of Thailand and the committee of the 1998 Asian Games to compose and conduct the music for opening ceremonies.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, COUNTRY, EVENT\n\nText:\nRohani was commissioned in 1998 by the government of Thailand and the committee of the 1998 Asian Games to compose and conduct the music for opening ceremonies.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_170", "text": "It has influenced, in some capacity, Garage house, Jungle music, Eurodance, Electropop, Dubstep, and even certain elements of Alternative rock and Hip hop.", "requested_types": ["MUSIC_GENRE", "SONG"], "distractors": ["SONG"], "gold": {"MUSIC_GENRE": ["Garage house", "Jungle music", "Eurodance", "Electropop", "Dubstep", "Alternative rock", "Hip hop"], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, SONG\n\nText:\nIt has influenced, in some capacity, Garage house, Jungle music, Eurodance, Electropop, Dubstep, and even certain elements of Alternative rock and Hip hop.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, SONG\n\nText:\nIt has influenced, in some capacity, Garage house, Jungle music, Eurodance, Electropop, Dubstep, and even certain elements of Alternative rock and Hip hop.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_171", "text": "Many early rock and roll songs are based on blues: That's All Right Mama, Johnny B. Goode, Blue Suede Shoes, Whole Lotta Shakin 'Goin On, Shake, Rattle, and Roll, and Long Tall Sally.", "requested_types": ["MISC", "SONG", "BAND", "MUSIC_GENRE"], "distractors": ["BAND", "MISC"], "gold": {"MISC": [], "SONG": ["That's All Right Mama", "Johnny B. Goode", "Blue Suede Shoes", "Whole Lotta Shakin 'Goin On", "Shake, Rattle, and Roll", "Long Tall Sally"], "BAND": [], "MUSIC_GENRE": ["rock and roll", "blues"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, BAND, MUSIC_GENRE\n\nText:\nMany early rock and roll songs are based on blues: That's All Right Mama, Johnny B. Goode, Blue Suede Shoes, Whole Lotta Shakin 'Goin On, Shake, Rattle, and Roll, and Long Tall Sally.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, BAND, MUSIC_GENRE\n\nText:\nMany early rock and roll songs are based on blues: That's All Right Mama, Johnny B. Goode, Blue Suede Shoes, Whole Lotta Shakin 'Goin On, Shake, Rattle, and Roll, and Long Tall Sally.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_172", "text": "The music video for the album's lead single, Virtual Insanity, was named MTV Video Music Award for Video of the Year at the 1997 MTV Video Music Awards.", "requested_types": ["MUSIC_GENRE", "SONG", "AWARD"], "distractors": ["MUSIC_GENRE"], "gold": {"MUSIC_GENRE": [], "SONG": ["Virtual Insanity"], "AWARD": ["MTV Video Music Award for Video of the Year", "1997 MTV Video Music Awards"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, SONG, AWARD\n\nText:\nThe music video for the album's lead single, Virtual Insanity, was named MTV Video Music Award for Video of the Year at the 1997 MTV Video Music 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: MUSIC_GENRE, SONG, AWARD\n\nText:\nThe music video for the album's lead single, Virtual Insanity, was named MTV Video Music Award for Video of the Year at the 1997 MTV Video Music 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_music_173", "text": "In June 2010, ASCAP sent letters to its members soliciting donations to fight entities that support weaker copyright restrictions, such as Public Knowledge, the Electronic Frontier Foundation, and Creative Commons,", "requested_types": ["ORGANISATION", "MUSIC_GENRE", "BAND", "MUSICAL_INSTRUMENT"], "distractors": ["MUSIC_GENRE", "MUSICAL_INSTRUMENT", "BAND"], "gold": {"ORGANISATION": ["ASCAP", "Public Knowledge", "Electronic Frontier Foundation", "Creative Commons"], "MUSIC_GENRE": [], "BAND": [], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, BAND, MUSICAL_INSTRUMENT\n\nText:\nIn June 2010, ASCAP sent letters to its members soliciting donations to fight entities that support weaker copyright restrictions, such as Public Knowledge, the Electronic Frontier Foundation, and Creative Commons,\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, BAND, MUSICAL_INSTRUMENT\n\nText:\nIn June 2010, ASCAP sent letters to its members soliciting donations to fight entities that support weaker copyright restrictions, such as Public Knowledge, the Electronic Frontier Foundation, and Creative Commons,\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_174", "text": "Hee Haw featured a premiere showcase on commercial television throughout its run for Country music, Bluegrass music, Gospel music, and other styles of American traditional music, featuring hundreds of elite musical performances that were paramount to the success, popularity and legacy of the series for a broad audience of Southern, rural and purely music fans alike.", "requested_types": ["SONG", "MUSIC_GENRE", "MISC", "COUNTRY"], "distractors": ["SONG", "COUNTRY"], "gold": {"SONG": [], "MUSIC_GENRE": ["Country music", "Bluegrass music", "Gospel music", "American traditional music", "Southern", "rural", "purely music"], "MISC": ["Hee Haw"], "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: SONG, MUSIC_GENRE, MISC, COUNTRY\n\nText:\nHee Haw featured a premiere showcase on commercial television throughout its run for Country music, Bluegrass music, Gospel music, and other styles of American traditional music, featuring hundreds of elite musical performances that were paramount to the success, popularity and legacy of the series for a broad audience of Southern, rural and purely music fans alike.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE, MISC, COUNTRY\n\nText:\nHee Haw featured a premiere showcase on commercial television throughout its run for Country music, Bluegrass music, Gospel music, and other styles of American traditional music, featuring hundreds of elite musical performances that were paramount to the success, popularity and legacy of the series for a broad audience of Southern, rural and purely music fans alike.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_175", "text": "In the early 2010s, there was somewhat of a resurgence of boy band popularity in countries where the trend had not maintained, with the emergence of new boy bands like Big Time Rush, The Wanted, and One Direction and the formation of supergroup NKOTBSB which comprised members of New Kids on the Block and Backstreet Boys.", "requested_types": ["EVENT", "BAND"], "distractors": ["EVENT"], "gold": {"EVENT": [], "BAND": ["Big Time Rush", "The Wanted", "One Direction", "NKOTBSB", "New Kids on the Block", "Backstreet Boys"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nIn the early 2010s, there was somewhat of a resurgence of boy band popularity in countries where the trend had not maintained, with the emergence of new boy bands like Big Time Rush, The Wanted, and One Direction and the formation of supergroup NKOTBSB which comprised members of New Kids on the Block and Backstreet Boys.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nIn the early 2010s, there was somewhat of a resurgence of boy band popularity in countries where the trend had not maintained, with the emergence of new boy bands like Big Time Rush, The Wanted, and One Direction and the formation of supergroup NKOTBSB which comprised members of New Kids on the Block and Backstreet Boys.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_176", "text": "The Pixies, minus original bassist Kim Deal, reunited for a United States and world tour in 2014 and have subsequently released three additional studio albums: Indie Cindy (2014), Head Carrier (2016) and Beneath the Eyrie (2019).", "requested_types": ["ALBUM", "LOCATION", "COUNTRY", "MUSICAL_ARTIST", "BAND"], "distractors": ["LOCATION"], "gold": {"ALBUM": ["Indie Cindy", "Head Carrier", "Beneath the Eyrie"], "LOCATION": [], "COUNTRY": ["United States"], "MUSICAL_ARTIST": ["Kim Deal"], "BAND": ["Pixies"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, LOCATION, COUNTRY, MUSICAL_ARTIST, BAND\n\nText:\nThe Pixies, minus original bassist Kim Deal, reunited for a United States and world tour in 2014 and have subsequently released three additional studio albums: Indie Cindy (2014), Head Carrier (2016) and Beneath the Eyrie (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: ALBUM, LOCATION, COUNTRY, MUSICAL_ARTIST, BAND\n\nText:\nThe Pixies, minus original bassist Kim Deal, reunited for a United States and world tour in 2014 and have subsequently released three additional studio albums: Indie Cindy (2014), Head Carrier (2016) and Beneath the Eyrie (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_music_177", "text": "Her most popular songs include All I Wanna Do (1993), If It Makes You Happy (1996), Everyday Is a Winding Road (1996), Tomorrow Never Dies (1997) (the theme song for the James Bond film of the same name), and My Favorite Mistake (1998).", "requested_types": ["PERSON", "MUSIC_GENRE", "SONG"], "distractors": ["MUSIC_GENRE"], "gold": {"PERSON": ["James Bond"], "MUSIC_GENRE": [], "SONG": ["All I Wanna Do", "If It Makes You Happy", "Everyday Is a Winding Road", "Tomorrow Never Dies", "My Favorite Mistake"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, SONG\n\nText:\nHer most popular songs include All I Wanna Do (1993), If It Makes You Happy (1996), Everyday Is a Winding Road (1996), Tomorrow Never Dies (1997) (the theme song for the James Bond film of the same name), and My Favorite Mistake (1998).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, SONG\n\nText:\nHer most popular songs include All I Wanna Do (1993), If It Makes You Happy (1996), Everyday Is a Winding Road (1996), Tomorrow Never Dies (1997) (the theme song for the James Bond film of the same name), and My Favorite Mistake (1998).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_178", "text": "He traveled to Africa in 1973, where he visited Ethiopia, Kenya, Tanzania, Malawi, and South Africa.", "requested_types": ["BAND", "LOCATION", "COUNTRY", "SONG"], "distractors": ["BAND", "SONG"], "gold": {"BAND": [], "LOCATION": ["Africa"], "COUNTRY": ["Ethiopia", "Kenya", "Tanzania", "Malawi", "South Africa"], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, LOCATION, COUNTRY, SONG\n\nText:\nHe traveled to Africa in 1973, where he visited Ethiopia, Kenya, Tanzania, Malawi, and South Africa.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, LOCATION, COUNTRY, SONG\n\nText:\nHe traveled to Africa in 1973, where he visited Ethiopia, Kenya, Tanzania, Malawi, and South Africa.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_179", "text": "Particularly noted in the UK are the Music of Newport, once labelled 'the new Seattle', and the Music of Cardiff, for which the city has recently been labelled 'Music City', for having the second highest number of independent music venues in the UK.", "requested_types": ["MUSICAL_INSTRUMENT", "MUSICAL_ARTIST", "COUNTRY", "BAND", "LOCATION"], "distractors": ["MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "BAND"], "gold": {"MUSICAL_INSTRUMENT": [], "MUSICAL_ARTIST": [], "COUNTRY": ["UK"], "BAND": [], "LOCATION": ["Music of Newport", "new Seattle", "Music of Cardiff", "Music City"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSICAL_ARTIST, COUNTRY, BAND, LOCATION\n\nText:\nParticularly noted in the UK are the Music of Newport, once labelled 'the new Seattle', and the Music of Cardiff, for which the city has recently been labelled 'Music City', for having the second highest number of independent music venues in the UK.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSICAL_ARTIST, COUNTRY, BAND, LOCATION\n\nText:\nParticularly noted in the UK are the Music of Newport, once labelled 'the new Seattle', and the Music of Cardiff, for which the city has recently been labelled 'Music City', for having the second highest number of independent music venues in the UK.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_180", "text": "Lords of Acid has also explored various other genres of music such as Hip hop music and R & amp; B, disco, Pop music, jazz, Latin, reggae, Folk music and Country music.", "requested_types": ["MUSIC_GENRE", "BAND", "PERSON", "AWARD", "LOCATION"], "distractors": ["LOCATION", "AWARD", "PERSON"], "gold": {"MUSIC_GENRE": ["Hip hop music", "R & amp; B", "disco", "Pop music", "jazz", "Latin", "reggae", "Folk music", "Country music"], "BAND": ["Lords of Acid"], "PERSON": [], "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: MUSIC_GENRE, BAND, PERSON, AWARD, LOCATION\n\nText:\nLords of Acid has also explored various other genres of music such as Hip hop music and R & amp; B, disco, Pop music, jazz, Latin, reggae, Folk music and Country music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, BAND, PERSON, AWARD, LOCATION\n\nText:\nLords of Acid has also explored various other genres of music such as Hip hop music and R & amp; B, disco, Pop music, jazz, Latin, reggae, Folk music and Country music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_181", "text": "In 2001, having previously won an Emmy Award, a Grammy Award and an Academy Awards, he joined a small list of EGOT winners with his Tony Award wins for The Producers.", "requested_types": ["SONG", "AWARD", "MISC"], "distractors": ["SONG"], "gold": {"SONG": [], "AWARD": ["Emmy Award", "Grammy Award", "Academy Awards", "EGOT", "Tony Award"], "MISC": ["The Producers"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, AWARD, MISC\n\nText:\nIn 2001, having previously won an Emmy Award, a Grammy Award and an Academy Awards, he joined a small list of EGOT winners with his Tony Award wins for The Producers.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, AWARD, MISC\n\nText:\nIn 2001, having previously won an Emmy Award, a Grammy Award and an Academy Awards, he joined a small list of EGOT winners with his Tony Award wins for The Producers.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_182", "text": "Steel-string guitars are also important in the world of flatpicking, as utilized by such artists as Clarence White, Tony Rice, Bryan Sutton, Doc Watson and David Grier.", "requested_types": ["AWARD", "SONG", "MUSICAL_INSTRUMENT", "MUSICAL_ARTIST", "MISC"], "distractors": ["AWARD", "SONG"], "gold": {"AWARD": [], "SONG": [], "MUSICAL_INSTRUMENT": ["Steel-string guitars"], "MUSICAL_ARTIST": ["Clarence White", "Tony Rice", "Bryan Sutton", "Doc Watson", "David Grier"], "MISC": ["flatpicking"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, MISC\n\nText:\nSteel-string guitars are also important in the world of flatpicking, as utilized by such artists as Clarence White, Tony Rice, Bryan Sutton, Doc Watson and David Grier.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, MISC\n\nText:\nSteel-string guitars are also important in the world of flatpicking, as utilized by such artists as Clarence White, Tony Rice, Bryan Sutton, Doc Watson and David Grier.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_183", "text": "In addition to Lady Antebellum, groups such as Herrick, The Quebe Sisters Band, Little Big Town, The Band Perry, Gloriana, Thompson Square, Eli Young Band, Zac Brown Band and British duo The Shires have emerged to occupy a large portion of the new country artists in the popular scene along with solo singers Kacey Musgraves and Miranda Lambert.", "requested_types": ["BAND", "MUSIC_GENRE", "MISC", "MUSICAL_ARTIST", "COUNTRY", "ALBUM"], "distractors": ["ALBUM", "COUNTRY", "MUSIC_GENRE"], "gold": {"BAND": ["Lady Antebellum", "Herrick", "The Quebe Sisters Band", "Little Big Town", "The Band Perry", "Gloriana", "Thompson Square", "Eli Young Band", "Zac Brown Band", "The Shires"], "MUSIC_GENRE": [], "MISC": ["British"], "MUSICAL_ARTIST": ["Kacey Musgraves", "Miranda Lambert"], "COUNTRY": [], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, MISC, MUSICAL_ARTIST, COUNTRY, ALBUM\n\nText:\nIn addition to Lady Antebellum, groups such as Herrick, The Quebe Sisters Band, Little Big Town, The Band Perry, Gloriana, Thompson Square, Eli Young Band, Zac Brown Band and British duo The Shires have emerged to occupy a large portion of the new country artists in the popular scene along with solo singers Kacey Musgraves and Miranda Lambert.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, MISC, MUSICAL_ARTIST, COUNTRY, ALBUM\n\nText:\nIn addition to Lady Antebellum, groups such as Herrick, The Quebe Sisters Band, Little Big Town, The Band Perry, Gloriana, Thompson Square, Eli Young Band, Zac Brown Band and British duo The Shires have emerged to occupy a large portion of the new country artists in the popular scene along with solo singers Kacey Musgraves and Miranda Lambert.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_184", "text": "Organised by the European Broadcasting Union (EBU) and host broadcaster Latvijas Televīzija (LTV), the contest was held at the Skonto Hall, with the final on 24 May 2003.", "requested_types": ["MUSICAL_ARTIST", "LOCATION", "ORGANISATION", "MISC"], "distractors": ["MISC", "MUSICAL_ARTIST"], "gold": {"MUSICAL_ARTIST": [], "LOCATION": ["Skonto Hall"], "ORGANISATION": ["European Broadcasting Union", "EBU", "Latvijas Televīzija", "LTV"], "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: MUSICAL_ARTIST, LOCATION, ORGANISATION, MISC\n\nText:\nOrganised by the European Broadcasting Union (EBU) and host broadcaster Latvijas Televīzija (LTV), the contest was held at the Skonto Hall, with the final on 24 May 2003.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, LOCATION, ORGANISATION, MISC\n\nText:\nOrganised by the European Broadcasting Union (EBU) and host broadcaster Latvijas Televīzija (LTV), the contest was held at the Skonto Hall, with the final on 24 May 2003.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_185", "text": "In February 1969, RCA released the live album Bless Its Pointed Little Head, which was culled from 1968 performances at the Fillmore West on October 24-26 and the Fillmore East on November 28-30.", "requested_types": ["ORGANISATION", "LOCATION", "MUSICAL_ARTIST", "ALBUM", "AWARD"], "distractors": ["AWARD", "MUSICAL_ARTIST"], "gold": {"ORGANISATION": ["RCA"], "LOCATION": ["Fillmore West", "Fillmore East"], "MUSICAL_ARTIST": [], "ALBUM": ["Bless Its Pointed Little Head"], "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, LOCATION, MUSICAL_ARTIST, ALBUM, AWARD\n\nText:\nIn February 1969, RCA released the live album Bless Its Pointed Little Head, which was culled from 1968 performances at the Fillmore West on October 24-26 and the Fillmore East on November 28-30.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ALBUM, AWARD\n\nText:\nIn February 1969, RCA released the live album Bless Its Pointed Little Head, which was culled from 1968 performances at the Fillmore West on October 24-26 and the Fillmore East on November 28-30.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_186", "text": "Seal has won multiple awards throughout his career, including three Brit Awards; he won Brit Award for British Male Solo Artist in 1992, as well as four Grammy Award s and an MTV Video Music Award.", "requested_types": ["MUSICAL_ARTIST", "AWARD", "ORGANISATION", "MUSIC_GENRE"], "distractors": ["MUSIC_GENRE", "ORGANISATION"], "gold": {"MUSICAL_ARTIST": ["Seal"], "AWARD": ["Brit Awards", "Brit Award for British Male Solo Artist", "Grammy Award", "MTV Video Music Award"], "ORGANISATION": [], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, AWARD, ORGANISATION, MUSIC_GENRE\n\nText:\nSeal has won multiple awards throughout his career, including three Brit Awards; he won Brit Award for British Male Solo Artist in 1992, as well as four Grammy Award s and an MTV Video Music 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: MUSICAL_ARTIST, AWARD, ORGANISATION, MUSIC_GENRE\n\nText:\nSeal has won multiple awards throughout his career, including three Brit Awards; he won Brit Award for British Male Solo Artist in 1992, as well as four Grammy Award s and an MTV Video Music 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_music_187", "text": "Four demos for the album were recorded on August 13, 1990; Enter Sandman, The Unforgiven, Nothing Else Matters and Wherever I May Roam.", "requested_types": ["SONG", "MUSIC_GENRE"], "distractors": ["MUSIC_GENRE"], "gold": {"SONG": ["Enter Sandman", "The Unforgiven", "Nothing Else Matters", "Wherever I May Roam"], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE\n\nText:\nFour demos for the album were recorded on August 13, 1990; Enter Sandman, The Unforgiven, Nothing Else Matters and Wherever I May Roam.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE\n\nText:\nFour demos for the album were recorded on August 13, 1990; Enter Sandman, The Unforgiven, Nothing Else Matters and Wherever I May Roam.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_188", "text": "In 2006, along with Sonu Nigam, Sunidhi Chauhan and Shiamak Davar, Ghoshal performed the theme song of 2010 Commonwealth Games at its closing ceremony, as an invitation to everyone to the following Commonwealth Games in Delhi.", "requested_types": ["EVENT", "MUSICAL_ARTIST", "MUSIC_GENRE", "LOCATION"], "distractors": ["MUSIC_GENRE"], "gold": {"EVENT": ["2010 Commonwealth Games", "Commonwealth Games"], "MUSICAL_ARTIST": ["Sonu Nigam", "Sunidhi Chauhan", "Shiamak Davar", "Ghoshal"], "MUSIC_GENRE": [], "LOCATION": ["Delhi"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, MUSIC_GENRE, LOCATION\n\nText:\nIn 2006, along with Sonu Nigam, Sunidhi Chauhan and Shiamak Davar, Ghoshal performed the theme song of 2010 Commonwealth Games at its closing ceremony, as an invitation to everyone to the following Commonwealth Games in Delhi.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, MUSIC_GENRE, LOCATION\n\nText:\nIn 2006, along with Sonu Nigam, Sunidhi Chauhan and Shiamak Davar, Ghoshal performed the theme song of 2010 Commonwealth Games at its closing ceremony, as an invitation to everyone to the following Commonwealth Games in Delhi.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_189", "text": "King Crimson 's In the Court of the Crimson King started with 21st Century Schizoid Man, which was considered heavy metal by several critics. Fricke David.", "requested_types": ["MUSICAL_INSTRUMENT", "AWARD", "PERSON", "MUSIC_GENRE", "BAND", "MISC", "ALBUM", "SONG"], "distractors": ["MISC", "MUSICAL_INSTRUMENT", "AWARD"], "gold": {"MUSICAL_INSTRUMENT": [], "AWARD": [], "PERSON": ["Fricke David"], "MUSIC_GENRE": ["heavy metal"], "BAND": ["King Crimson"], "MISC": [], "ALBUM": ["In the Court of the Crimson King"], "SONG": ["21st Century Schizoid Man"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, AWARD, PERSON, MUSIC_GENRE, BAND, MISC, ALBUM, SONG\n\nText:\nKing Crimson 's In the Court of the Crimson King started with 21st Century Schizoid Man, which was considered heavy metal by several critics. Fricke David.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, AWARD, PERSON, MUSIC_GENRE, BAND, MISC, ALBUM, SONG\n\nText:\nKing Crimson 's In the Court of the Crimson King started with 21st Century Schizoid Man, which was considered heavy metal by several critics. Fricke David.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_190", "text": "She reprised her role in it the next year, playing the Edinburgh Playhouse from November 19 to December 8, 2007 and the Wales Millennium Centre in the Donald Gordon Theatre from December 13, 2007 through January 12, 2008.", "requested_types": ["LOCATION", "ORGANISATION"], "distractors": ["ORGANISATION"], "gold": {"LOCATION": ["Edinburgh Playhouse", "Wales Millennium Centre", "Donald Gordon Theatre"], "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: LOCATION, ORGANISATION\n\nText:\nShe reprised her role in it the next year, playing the Edinburgh Playhouse from November 19 to December 8, 2007 and the Wales Millennium Centre in the Donald Gordon Theatre from December 13, 2007 through January 12, 2008.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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\n\nText:\nShe reprised her role in it the next year, playing the Edinburgh Playhouse from November 19 to December 8, 2007 and the Wales Millennium Centre in the Donald Gordon Theatre from December 13, 2007 through January 12, 2008.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_191", "text": "He is also a recipient of the AFI Life Achievement Award for his contributions to the cinema, and has won an Academy Awards, a Palme d'Or, Cannes Film Festival Best Director Award, Silver Lion, Grammy Award, Emmy Award, Golden Globes, British Academy Film Awards, and Directors Guild of America Award s.", "requested_types": ["AWARD", "EVENT"], "distractors": ["EVENT"], "gold": {"AWARD": ["AFI Life Achievement Award", "Academy Awards", "Palme d'Or", "Cannes Film Festival Best Director Award", "Silver Lion", "Grammy Award", "Emmy Award", "Golden Globes", "British Academy Film Awards", "Directors Guild of America 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: AWARD, EVENT\n\nText:\nHe is also a recipient of the AFI Life Achievement Award for his contributions to the cinema, and has won an Academy Awards, a Palme d'Or, Cannes Film Festival Best Director Award, Silver Lion, Grammy Award, Emmy Award, Golden Globes, British Academy Film Awards, and Directors Guild of America Award 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: AWARD, EVENT\n\nText:\nHe is also a recipient of the AFI Life Achievement Award for his contributions to the cinema, and has won an Academy Awards, a Palme d'Or, Cannes Film Festival Best Director Award, Silver Lion, Grammy Award, Emmy Award, Golden Globes, British Academy Film Awards, and Directors Guild of America Award 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_music_192", "text": "With the advent of the radio, bugle signaling units became obsolete and surplus equipment was sold to veteran organizations (such as the Veterans of Foreign Wars and American Legion, two major organizers for classic drum corps).", "requested_types": ["ORGANISATION", "MUSICAL_INSTRUMENT", "MUSICAL_ARTIST", "MISC"], "distractors": ["MUSICAL_ARTIST", "MUSICAL_INSTRUMENT"], "gold": {"ORGANISATION": ["Veterans of Foreign Wars", "American Legion"], "MUSICAL_INSTRUMENT": [], "MUSICAL_ARTIST": [], "MISC": ["radio", "classic drum corps"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, MISC\n\nText:\nWith the advent of the radio, bugle signaling units became obsolete and surplus equipment was sold to veteran organizations (such as the Veterans of Foreign Wars and American Legion, two major organizers for classic drum corps).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, MISC\n\nText:\nWith the advent of the radio, bugle signaling units became obsolete and surplus equipment was sold to veteran organizations (such as the Veterans of Foreign Wars and American Legion, two major organizers for classic drum corps).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_193", "text": "A few of those include bagad (Breton pipe bands), Fairport Convention, Pentangle, Steeleye Span and Horslips.", "requested_types": ["EVENT", "BAND"], "distractors": ["EVENT"], "gold": {"EVENT": [], "BAND": ["bagad", "Breton pipe bands", "Fairport Convention", "Pentangle", "Steeleye Span", "Horslips"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nA few of those include bagad (Breton pipe bands), Fairport Convention, Pentangle, Steeleye Span and Horslips.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nA few of those include bagad (Breton pipe bands), Fairport Convention, Pentangle, Steeleye Span and Horslips.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_194", "text": "Bubbles created sleeves for bands including the Damned, Elvis Costello, Ian Dury and Wreckless Eric.", "requested_types": ["BAND", "AWARD", "EVENT", "PERSON", "MUSICAL_ARTIST", "COUNTRY"], "distractors": ["AWARD", "COUNTRY", "EVENT"], "gold": {"BAND": ["the Damned"], "AWARD": [], "EVENT": [], "PERSON": ["Bubbles"], "MUSICAL_ARTIST": ["Elvis Costello", "Ian Dury", "Wreckless Eric"], "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: BAND, AWARD, EVENT, PERSON, MUSICAL_ARTIST, COUNTRY\n\nText:\nBubbles created sleeves for bands including the Damned, Elvis Costello, Ian Dury and Wreckless Eric.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, AWARD, EVENT, PERSON, MUSICAL_ARTIST, COUNTRY\n\nText:\nBubbles created sleeves for bands including the Damned, Elvis Costello, Ian Dury and Wreckless Eric.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_195", "text": "Notable post-punk groups that presaged that genre and helped develop and shape the subculture, include Siouxsie and the Banshees, Joy Division, Bauhaus and The Cure.", "requested_types": ["BAND", "MUSIC_GENRE", "MISC", "ORGANISATION", "SONG"], "distractors": ["ORGANISATION", "MISC", "SONG"], "gold": {"BAND": ["Siouxsie and the Banshees", "Joy Division", "Bauhaus", "The Cure"], "MUSIC_GENRE": ["post-punk"], "MISC": [], "ORGANISATION": [], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, MISC, ORGANISATION, SONG\n\nText:\nNotable post-punk groups that presaged that genre and helped develop and shape the subculture, include Siouxsie and the Banshees, Joy Division, Bauhaus and The Cure.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, MISC, ORGANISATION, SONG\n\nText:\nNotable post-punk groups that presaged that genre and helped develop and shape the subculture, include Siouxsie and the Banshees, Joy Division, Bauhaus and The Cure.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_196", "text": "Jagger, in the role of Turner in the 1970 film Performance, performed excerpts from Come On in My Kitchen and Me and the Devil Blues.", "requested_types": ["MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "PERSON", "AWARD", "SONG", "MISC"], "distractors": ["AWARD", "MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_ARTIST": ["Jagger"], "MUSICAL_INSTRUMENT": [], "PERSON": ["Turner"], "AWARD": [], "SONG": ["Come On in My Kitchen", "Me and the Devil Blues"], "MISC": ["Performance"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MUSICAL_INSTRUMENT, PERSON, AWARD, SONG, MISC\n\nText:\nJagger, in the role of Turner in the 1970 film Performance, performed excerpts from Come On in My Kitchen and Me and the Devil Blues.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MUSICAL_INSTRUMENT, PERSON, AWARD, SONG, MISC\n\nText:\nJagger, in the role of Turner in the 1970 film Performance, performed excerpts from Come On in My Kitchen and Me and the Devil Blues.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_197", "text": "Among the most prominent of these were Metallica 's Master of Puppets, Slayer' s Reign in Blood, Anthrax 's Among the Living and Megadeth' s Peace Sells... but Who's Buying? beginning a search for his replacement.", "requested_types": ["COUNTRY", "BAND", "ALBUM"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "BAND": ["Metallica", "Slayer", "Anthrax", "Megadeth"], "ALBUM": ["Master of Puppets", "Reign in Blood", "Among the Living", "Peace Sells... but Who's Buying?"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM\n\nText:\nAmong the most prominent of these were Metallica 's Master of Puppets, Slayer' s Reign in Blood, Anthrax 's Among the Living and Megadeth' s Peace Sells... but Who's Buying? beginning a search for his replacement.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM\n\nText:\nAmong the most prominent of these were Metallica 's Master of Puppets, Slayer' s Reign in Blood, Anthrax 's Among the Living and Megadeth' s Peace Sells... but Who's Buying? beginning a search for his replacement.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_198", "text": "After a long campaign, the World DanceSport Federation (WDSF), formerly IDSF, was recognized by the International Olympic Committee as the sole representative body for dancesport, on September 5, 1997.Long, Daniel 1999.", "requested_types": ["MUSICAL_INSTRUMENT", "ORGANISATION"], "distractors": ["MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_INSTRUMENT": [], "ORGANISATION": ["World DanceSport Federation", "WDSF", "IDSF", "International Olympic Committee"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, ORGANISATION\n\nText:\nAfter a long campaign, the World DanceSport Federation (WDSF), formerly IDSF, was recognized by the International Olympic Committee as the sole representative body for dancesport, on September 5, 1997.Long, Daniel 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: MUSICAL_INSTRUMENT, ORGANISATION\n\nText:\nAfter a long campaign, the World DanceSport Federation (WDSF), formerly IDSF, was recognized by the International Olympic Committee as the sole representative body for dancesport, on September 5, 1997.Long, Daniel 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_music_199", "text": "Her music has been heard at the Brooklyn Academy of Music, the Sydney Olympic Arts Festival, Settembre Musica (Italy), Théâtre de la Ville (Paris), Lincoln Center for the Performing Arts, and Carnegie Hall, and has been recorded on Cantaloupe Music, Teldec, Point / Universal, Sony Classical Records, and Argo / Decca.", "requested_types": ["ORGANISATION", "ALBUM", "LOCATION", "EVENT", "COUNTRY"], "distractors": ["ALBUM"], "gold": {"ORGANISATION": ["Cantaloupe Music", "Teldec", "Point / Universal", "Sony Classical Records", "Argo / Decca"], "ALBUM": [], "LOCATION": ["Brooklyn Academy of Music", "Settembre Musica", "Théâtre de la Ville", "Paris", "Lincoln Center for the Performing Arts", "Carnegie Hall"], "EVENT": ["Sydney Olympic Arts Festival"], "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: ORGANISATION, ALBUM, LOCATION, EVENT, COUNTRY\n\nText:\nHer music has been heard at the Brooklyn Academy of Music, the Sydney Olympic Arts Festival, Settembre Musica (Italy), Théâtre de la Ville (Paris), Lincoln Center for the Performing Arts, and Carnegie Hall, and has been recorded on Cantaloupe Music, Teldec, Point / Universal, Sony Classical Records, and Argo / Decca.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, LOCATION, EVENT, COUNTRY\n\nText:\nHer music has been heard at the Brooklyn Academy of Music, the Sydney Olympic Arts Festival, Settembre Musica (Italy), Théâtre de la Ville (Paris), Lincoln Center for the Performing Arts, and Carnegie Hall, and has been recorded on Cantaloupe Music, Teldec, Point / Universal, Sony Classical Records, and Argo / Decca.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_200", "text": "Additionally, Grant won the Golden Globe Award for Best Actor - Motion Picture Musical or Comedy and the BAFTA Award for Best Actor in a Leading Role, and the film won the British Academy Film Awards BAFTA Award for Best Film, BAFTA Award for Best Direction, and BAFTA Award for Best Actress in a Supporting Role for Scott Thomas.", "requested_types": ["LOCATION", "SONG", "PERSON", "COUNTRY", "AWARD"], "distractors": ["SONG", "COUNTRY", "LOCATION"], "gold": {"LOCATION": [], "SONG": [], "PERSON": ["Grant", "Scott Thomas"], "COUNTRY": [], "AWARD": ["Golden Globe Award for Best Actor - Motion Picture Musical or Comedy", "BAFTA Award for Best Actor in a Leading Role", "British Academy Film Awards", "BAFTA Award for Best Film", "BAFTA Award for Best Direction", "BAFTA Award for Best Actress in a Supporting Role"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, PERSON, COUNTRY, AWARD\n\nText:\nAdditionally, Grant won the Golden Globe Award for Best Actor - Motion Picture Musical or Comedy and the BAFTA Award for Best Actor in a Leading Role, and the film won the British Academy Film Awards BAFTA Award for Best Film, BAFTA Award for Best Direction, and BAFTA Award for Best Actress in a Supporting Role for Scott Thomas.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, PERSON, COUNTRY, AWARD\n\nText:\nAdditionally, Grant won the Golden Globe Award for Best Actor - Motion Picture Musical or Comedy and the BAFTA Award for Best Actor in a Leading Role, and the film won the British Academy Film Awards BAFTA Award for Best Film, BAFTA Award for Best Direction, and BAFTA Award for Best Actress in a Supporting Role for Scott Thomas.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_201", "text": "XXXVII No. 32, pp. 69, 82) Spaces in Manhattan that supported Downtown music from the 1960s on included the Judson Memorial Church, The Kitchen, Experimental Intermedia, Roulette Intermedium, the Knitting Factory, Dance Theater Workshop, Tonic, the Gas Station, the Paula Cooper Gallery, and others.", "requested_types": ["LOCATION", "MUSIC_GENRE", "EVENT"], "distractors": ["EVENT"], "gold": {"LOCATION": ["Manhattan", "Judson Memorial Church", "The Kitchen", "Experimental Intermedia", "Roulette Intermedium", "Knitting Factory", "Dance Theater Workshop", "Tonic", "Gas Station", "Paula Cooper Gallery"], "MUSIC_GENRE": ["Downtown music"], "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: LOCATION, MUSIC_GENRE, EVENT\n\nText:\nXXXVII No. 32, pp. 69, 82) Spaces in Manhattan that supported Downtown music from the 1960s on included the Judson Memorial Church, The Kitchen, Experimental Intermedia, Roulette Intermedium, the Knitting Factory, Dance Theater Workshop, Tonic, the Gas Station, the Paula Cooper Gallery, 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: LOCATION, MUSIC_GENRE, EVENT\n\nText:\nXXXVII No. 32, pp. 69, 82) Spaces in Manhattan that supported Downtown music from the 1960s on included the Judson Memorial Church, The Kitchen, Experimental Intermedia, Roulette Intermedium, the Knitting Factory, Dance Theater Workshop, Tonic, the Gas Station, the Paula Cooper Gallery, 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_music_202", "text": "The singles One Night in Bangkok (with vocals by Murray Head and Anders Glenmark) and I Know Him So Well (a duet by Barbara Dickson and Elaine Paige, and later also recorded by Barbra Streisand and Whitney Houston) were both hugely successful.", "requested_types": ["MUSIC_GENRE", "MUSICAL_ARTIST", "SONG"], "distractors": ["MUSIC_GENRE"], "gold": {"MUSIC_GENRE": [], "MUSICAL_ARTIST": ["Murray Head", "Anders Glenmark", "Barbara Dickson", "Elaine Paige", "Barbra Streisand", "Whitney Houston"], "SONG": ["One Night in Bangkok", "I Know Him So Well"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, SONG\n\nText:\nThe singles One Night in Bangkok (with vocals by Murray Head and Anders Glenmark) and I Know Him So Well (a duet by Barbara Dickson and Elaine Paige, and later also recorded by Barbra Streisand and Whitney Houston) were both hugely successful.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, SONG\n\nText:\nThe singles One Night in Bangkok (with vocals by Murray Head and Anders Glenmark) and I Know Him So Well (a duet by Barbara Dickson and Elaine Paige, and later also recorded by Barbra Streisand and Whitney Houston) were both hugely successful.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_203", "text": "They took up a long residency at San Francisco's Black Hawk nightclub and gained great popularity touring college campuses, recording a series of albums with such titles as Jazz at Oberlin (1953), Jazz at the College of the Pacific (1953), and Brubeck's debut on Columbia Records, Jazz Goes to College (1954).", "requested_types": ["MUSICAL_ARTIST", "ALBUM", "ORGANISATION", "MUSIC_GENRE", "LOCATION"], "distractors": ["MUSIC_GENRE"], "gold": {"MUSICAL_ARTIST": ["Brubeck"], "ALBUM": ["Jazz at Oberlin", "Jazz at the College of the Pacific", "Jazz Goes to College"], "ORGANISATION": ["Columbia Records"], "MUSIC_GENRE": [], "LOCATION": ["San Francisco", "Black Hawk"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ALBUM, ORGANISATION, MUSIC_GENRE, LOCATION\n\nText:\nThey took up a long residency at San Francisco's Black Hawk nightclub and gained great popularity touring college campuses, recording a series of albums with such titles as Jazz at Oberlin (1953), Jazz at the College of the Pacific (1953), and Brubeck's debut on Columbia Records, Jazz Goes to College (1954).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ALBUM, ORGANISATION, MUSIC_GENRE, LOCATION\n\nText:\nThey took up a long residency at San Francisco's Black Hawk nightclub and gained great popularity touring college campuses, recording a series of albums with such titles as Jazz at Oberlin (1953), Jazz at the College of the Pacific (1953), and Brubeck's debut on Columbia Records, Jazz Goes to College (1954).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_204", "text": "Rollins left to join them soon after, leaving Davis to pay over $25,000 (US $With this group, Davis completed the rest of what became Seven Steps to Heaven (1963) and recorded the live albums Miles Davis in Europe (1964), My Funny Valentine (1965), and Four & More (1966).", "requested_types": ["ALBUM", "SONG", "MUSICAL_INSTRUMENT", "MUSICAL_ARTIST", "AWARD", "COUNTRY"], "distractors": ["MUSICAL_INSTRUMENT", "AWARD"], "gold": {"ALBUM": ["Seven Steps to Heaven", "Miles Davis in Europe", "Four & More"], "SONG": ["My Funny Valentine"], "MUSICAL_INSTRUMENT": [], "MUSICAL_ARTIST": ["Rollins", "Davis"], "AWARD": [], "COUNTRY": ["US"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, SONG, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, AWARD, COUNTRY\n\nText:\nRollins left to join them soon after, leaving Davis to pay over $25,000 (US $With this group, Davis completed the rest of what became Seven Steps to Heaven (1963) and recorded the live albums Miles Davis in Europe (1964), My Funny Valentine (1965), and Four & More (1966).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, SONG, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, AWARD, COUNTRY\n\nText:\nRollins left to join them soon after, leaving Davis to pay over $25,000 (US $With this group, Davis completed the rest of what became Seven Steps to Heaven (1963) and recorded the live albums Miles Davis in Europe (1964), My Funny Valentine (1965), and Four & More (1966).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_205", "text": "Bush's albums include the aforementioned Sixteen Stone, Razorblade Suitcase, Deconstructed, The Science of Things, and Golden State.", "requested_types": ["ALBUM", "EVENT", "MUSICAL_INSTRUMENT", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_INSTRUMENT", "EVENT"], "gold": {"ALBUM": ["Sixteen Stone", "Razorblade Suitcase", "Deconstructed", "The Science of Things", "Golden State"], "EVENT": [], "MUSICAL_INSTRUMENT": [], "MUSICAL_ARTIST": ["Bush"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, EVENT, MUSICAL_INSTRUMENT, MUSICAL_ARTIST\n\nText:\nBush's albums include the aforementioned Sixteen Stone, Razorblade Suitcase, Deconstructed, The Science of Things, and Golden State.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, EVENT, MUSICAL_INSTRUMENT, MUSICAL_ARTIST\n\nText:\nBush's albums include the aforementioned Sixteen Stone, Razorblade Suitcase, Deconstructed, The Science of Things, and Golden State.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_206", "text": "Douglas also began to appear at the mainstream jazz clubs around New York such as Iridium Jazz Club, Village Vanguard and Jazz Standard.", "requested_types": ["LOCATION", "MUSIC_GENRE", "MUSICAL_ARTIST", "ORGANISATION"], "distractors": ["ORGANISATION"], "gold": {"LOCATION": ["New York", "Iridium Jazz Club", "Village Vanguard", "Jazz Standard"], "MUSIC_GENRE": ["jazz"], "MUSICAL_ARTIST": ["Douglas"], "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: LOCATION, MUSIC_GENRE, MUSICAL_ARTIST, ORGANISATION\n\nText:\nDouglas also began to appear at the mainstream jazz clubs around New York such as Iridium Jazz Club, Village Vanguard and Jazz Standard.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, MUSICAL_ARTIST, ORGANISATION\n\nText:\nDouglas also began to appear at the mainstream jazz clubs around New York such as Iridium Jazz Club, Village Vanguard and Jazz Standard.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_207", "text": "For Revolver he selected five non- Heavy metal music records that influenced him: The Cure's Pornography, Helium's No Guitars, Mogwai EP + 2, My Bloody Valentine's Loveless and The Smashing Pumpkins' Siamese Dream.", "requested_types": ["ALBUM", "EVENT", "BAND", "SONG", "MUSIC_GENRE"], "distractors": ["SONG", "EVENT"], "gold": {"ALBUM": ["Pornography", "No Guitars", "EP + 2", "Loveless", "Siamese Dream"], "EVENT": [], "BAND": ["Revolver", "Cure", "Helium", "Mogwai", "My Bloody Valentine", "The Smashing Pumpkins"], "SONG": [], "MUSIC_GENRE": ["Heavy metal music"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, EVENT, BAND, SONG, MUSIC_GENRE\n\nText:\nFor Revolver he selected five non- Heavy metal music records that influenced him: The Cure's Pornography, Helium's No Guitars, Mogwai EP + 2, My Bloody Valentine's Loveless and The Smashing Pumpkins' Siamese Dream.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, EVENT, BAND, SONG, MUSIC_GENRE\n\nText:\nFor Revolver he selected five non- Heavy metal music records that influenced him: The Cure's Pornography, Helium's No Guitars, Mogwai EP + 2, My Bloody Valentine's Loveless and The Smashing Pumpkins' Siamese Dream.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_208", "text": "He has also released three solo albums: Stranger in This Town in 1991, Undiscovered Soul in 1998, and Aftermath of the Lowdown released in September 2012.", "requested_types": ["MUSIC_GENRE", "MISC", "ALBUM"], "distractors": ["MISC", "MUSIC_GENRE"], "gold": {"MUSIC_GENRE": [], "MISC": [], "ALBUM": ["Stranger in This Town", "Undiscovered Soul", "Aftermath of the Lowdown"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MISC, ALBUM\n\nText:\nHe has also released three solo albums: Stranger in This Town in 1991, Undiscovered Soul in 1998, and Aftermath of the Lowdown released in September 2012.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MISC, ALBUM\n\nText:\nHe has also released three solo albums: Stranger in This Town in 1991, Undiscovered Soul in 1998, and Aftermath of the Lowdown released in September 2012.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_209", "text": "He also tried his hand at music production, producing Idoli 's 1980 seven-inch single Maljčiki / Retko te viđam sa devojkama and co-producing, alongside Kornelije Kovač, Zdravko Čolić' s fourth studio album Malo pojačaj radio in 1981.", "requested_types": ["ALBUM", "AWARD", "COUNTRY", "MUSICAL_ARTIST", "SONG", "EVENT", "BAND"], "distractors": ["AWARD", "EVENT", "COUNTRY"], "gold": {"ALBUM": ["Malo pojačaj radio"], "AWARD": [], "COUNTRY": [], "MUSICAL_ARTIST": ["Kornelije Kovač", "Zdravko Čolić"], "SONG": ["Maljčiki", "Retko te viđam sa devojkama"], "EVENT": [], "BAND": ["Idoli"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, AWARD, COUNTRY, MUSICAL_ARTIST, SONG, EVENT, BAND\n\nText:\nHe also tried his hand at music production, producing Idoli 's 1980 seven-inch single Maljčiki / Retko te viđam sa devojkama and co-producing, alongside Kornelije Kovač, Zdravko Čolić' s fourth studio album Malo pojačaj radio in 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: ALBUM, AWARD, COUNTRY, MUSICAL_ARTIST, SONG, EVENT, BAND\n\nText:\nHe also tried his hand at music production, producing Idoli 's 1980 seven-inch single Maljčiki / Retko te viđam sa devojkama and co-producing, alongside Kornelije Kovač, Zdravko Čolić' s fourth studio album Malo pojačaj radio in 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_music_210", "text": "Follow-up albums You've Come a Long Way, Baby, Halfway Between the Gutter and the Stars, and Palookaville, as well as singles such as The Rockafeller Skank, Praise You, Right Here, Right Now, Weapon of Choice, and Wonderful Night, achieved commercial and critical success.", "requested_types": ["ALBUM", "LOCATION", "SONG"], "distractors": ["LOCATION"], "gold": {"ALBUM": ["You've Come a Long Way, Baby", "Halfway Between the Gutter and the Stars", "Palookaville"], "LOCATION": [], "SONG": ["The Rockafeller Skank", "Praise You", "Right Here, Right Now", "Weapon of Choice", "Wonderful Night"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, LOCATION, SONG\n\nText:\nFollow-up albums You've Come a Long Way, Baby, Halfway Between the Gutter and the Stars, and Palookaville, as well as singles such as The Rockafeller Skank, Praise You, Right Here, Right Now, Weapon of Choice, and Wonderful Night, achieved commercial and critical success.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, LOCATION, SONG\n\nText:\nFollow-up albums You've Come a Long Way, Baby, Halfway Between the Gutter and the Stars, and Palookaville, as well as singles such as The Rockafeller Skank, Praise You, Right Here, Right Now, Weapon of Choice, and Wonderful Night, achieved commercial and critical success.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_211", "text": "In 2002, Grohl helped Chan Marshall of Cat Power on the album You Are Free and played with Queens of the Stone Age on their album Songs for the Deaf.", "requested_types": ["BAND", "ALBUM", "SONG", "MUSICAL_ARTIST"], "distractors": ["SONG"], "gold": {"BAND": ["Queens of the Stone Age"], "ALBUM": ["You Are Free", "Songs for the Deaf"], "SONG": [], "MUSICAL_ARTIST": ["Grohl", "Chan Marshall", "Cat Power"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, ALBUM, SONG, MUSICAL_ARTIST\n\nText:\nIn 2002, Grohl helped Chan Marshall of Cat Power on the album You Are Free and played with Queens of the Stone Age on their album Songs for the Deaf.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, ALBUM, SONG, MUSICAL_ARTIST\n\nText:\nIn 2002, Grohl helped Chan Marshall of Cat Power on the album You Are Free and played with Queens of the Stone Age on their album Songs for the Deaf.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_212", "text": "He recorded four studio albums for A & amp; M: Pleasures of the Harbor (1967), Tape from California (1968), Rehearsals for Retirement (1969), and the ironically titled Greatest Hits (1970) (which actually consisted of all new material).", "requested_types": ["ORGANISATION", "ALBUM", "PERSON", "LOCATION"], "distractors": ["PERSON", "LOCATION"], "gold": {"ORGANISATION": ["A & amp; M"], "ALBUM": ["Pleasures of the Harbor", "Tape from California", "Rehearsals for Retirement", "Greatest Hits"], "PERSON": [], "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, ALBUM, PERSON, LOCATION\n\nText:\nHe recorded four studio albums for A & amp; M: Pleasures of the Harbor (1967), Tape from California (1968), Rehearsals for Retirement (1969), and the ironically titled Greatest Hits (1970) (which actually consisted of all new material).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, PERSON, LOCATION\n\nText:\nHe recorded four studio albums for A & amp; M: Pleasures of the Harbor (1967), Tape from California (1968), Rehearsals for Retirement (1969), and the ironically titled Greatest Hits (1970) (which actually consisted of all new material).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_213", "text": "Howlin 'Wolf, Muddy Waters, Willie Dixon and Jimmy Reed were all born in Mississippi and moved to Chicago during the Great Migration.", "requested_types": ["MISC", "LOCATION", "BAND", "EVENT", "MUSICAL_ARTIST"], "distractors": ["BAND", "MISC"], "gold": {"MISC": [], "LOCATION": ["Mississippi", "Chicago"], "BAND": [], "EVENT": ["Great Migration"], "MUSICAL_ARTIST": ["Howlin 'Wolf", "Muddy Waters", "Willie Dixon", "Jimmy Reed"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, EVENT, MUSICAL_ARTIST\n\nText:\nHowlin 'Wolf, Muddy Waters, Willie Dixon and Jimmy Reed were all born in Mississippi and moved to Chicago during the Great Migration.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, EVENT, MUSICAL_ARTIST\n\nText:\nHowlin 'Wolf, Muddy Waters, Willie Dixon and Jimmy Reed were all born in Mississippi and moved to Chicago during the Great Migration.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_214", "text": "It received a major expansion ahead of the 1983 Summer Universiade, when it reached a capacity of 60,081.", "requested_types": ["AWARD", "EVENT"], "distractors": ["AWARD"], "gold": {"AWARD": [], "EVENT": ["1983 Summer Universiade"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT\n\nText:\nIt received a major expansion ahead of the 1983 Summer Universiade, when it reached a capacity of 60,081.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT\n\nText:\nIt received a major expansion ahead of the 1983 Summer Universiade, when it reached a capacity of 60,081.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_215", "text": "Other artists influenced by McTell include Taj Mahal, Alvin Youngblood Hart, Ralph McTell, Chris Smither, Jack White, and the The White Stripes.", "requested_types": ["MUSIC_GENRE", "MUSICAL_ARTIST", "BAND"], "distractors": ["MUSIC_GENRE"], "gold": {"MUSIC_GENRE": [], "MUSICAL_ARTIST": ["McTell", "Taj Mahal", "Alvin Youngblood Hart", "Ralph McTell", "Chris Smither", "Jack White"], "BAND": ["The White Stripes"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, BAND\n\nText:\nOther artists influenced by McTell include Taj Mahal, Alvin Youngblood Hart, Ralph McTell, Chris Smither, Jack White, and the The White Stripes.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, BAND\n\nText:\nOther artists influenced by McTell include Taj Mahal, Alvin Youngblood Hart, Ralph McTell, Chris Smither, Jack White, and the The White Stripes.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_216", "text": "Reggae en Español spread from the Spanish-speaking Central American country of Panama to the mainland South American countries of Venezuela and Guyana then to the rest of South America.", "requested_types": ["MUSIC_GENRE", "COUNTRY", "ORGANISATION", "LOCATION", "MISC"], "distractors": ["MISC", "ORGANISATION"], "gold": {"MUSIC_GENRE": ["Reggae en Español"], "COUNTRY": ["Panama", "Venezuela", "Guyana"], "ORGANISATION": [], "LOCATION": ["Central American", "South American", "South America"], "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: MUSIC_GENRE, COUNTRY, ORGANISATION, LOCATION, MISC\n\nText:\nReggae en Español spread from the Spanish-speaking Central American country of Panama to the mainland South American countries of Venezuela and Guyana then to the rest of South 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: MUSIC_GENRE, COUNTRY, ORGANISATION, LOCATION, MISC\n\nText:\nReggae en Español spread from the Spanish-speaking Central American country of Panama to the mainland South American countries of Venezuela and Guyana then to the rest of South 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_music_217", "text": "Cliff Richard covered Lay All Your Love on Me, while Dionne Warwick, Peter Cetera, and Celebrity Skin recorded their versions of SOS.", "requested_types": ["SONG", "MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "LOCATION", "ALBUM", "BAND"], "distractors": ["LOCATION", "BAND", "MUSICAL_INSTRUMENT"], "gold": {"SONG": ["Lay All Your Love on Me", "SOS"], "MUSICAL_ARTIST": ["Cliff Richard", "Dionne Warwick", "Peter Cetera"], "MUSICAL_INSTRUMENT": [], "LOCATION": [], "ALBUM": ["Celebrity Skin"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, LOCATION, ALBUM, BAND\n\nText:\nCliff Richard covered Lay All Your Love on Me, while Dionne Warwick, Peter Cetera, and Celebrity Skin recorded their versions of SOS.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, LOCATION, ALBUM, BAND\n\nText:\nCliff Richard covered Lay All Your Love on Me, while Dionne Warwick, Peter Cetera, and Celebrity Skin recorded their versions of SOS.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_218", "text": "In 2003, Afshin-Jam became Miss World Canada and joined in the Miss World contest in Sanya, China, where she placed second.", "requested_types": ["ORGANISATION", "PERSON", "MUSICAL_INSTRUMENT", "MUSIC_GENRE", "COUNTRY", "SONG", "LOCATION"], "distractors": ["MUSICAL_INSTRUMENT", "MUSIC_GENRE", "SONG"], "gold": {"ORGANISATION": ["Miss World Canada", "Miss World"], "PERSON": ["Afshin-Jam"], "MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": [], "COUNTRY": ["China"], "SONG": [], "LOCATION": ["Sanya"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, MUSIC_GENRE, COUNTRY, SONG, LOCATION\n\nText:\nIn 2003, Afshin-Jam became Miss World Canada and joined in the Miss World contest in Sanya, China, where she placed second.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, MUSIC_GENRE, COUNTRY, SONG, LOCATION\n\nText:\nIn 2003, Afshin-Jam became Miss World Canada and joined in the Miss World contest in Sanya, China, where she placed second.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_219", "text": "On June 11-12, 2008, they played in Philadelphia and New Jersey at the Trocadero Theatre and Starland Ballroom, respectively.", "requested_types": ["LOCATION", "ALBUM", "ORGANISATION"], "distractors": ["ALBUM", "ORGANISATION"], "gold": {"LOCATION": ["Philadelphia", "New Jersey", "Trocadero Theatre", "Starland Ballroom"], "ALBUM": [], "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: LOCATION, ALBUM, ORGANISATION\n\nText:\nOn June 11-12, 2008, they played in Philadelphia and New Jersey at the Trocadero Theatre and Starland Ballroom, 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: LOCATION, ALBUM, ORGANISATION\n\nText:\nOn June 11-12, 2008, they played in Philadelphia and New Jersey at the Trocadero Theatre and Starland Ballroom, 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_music_220", "text": "Brad Wilk of Rage Against the Machine was chosen as the drummer, and Rick Rubin was chosen as the producer.", "requested_types": ["EVENT", "BAND", "MUSICAL_ARTIST", "AWARD"], "distractors": ["EVENT", "AWARD"], "gold": {"EVENT": [], "BAND": ["Rage Against the Machine"], "MUSICAL_ARTIST": ["Brad Wilk", "Rick Rubin"], "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: EVENT, BAND, MUSICAL_ARTIST, AWARD\n\nText:\nBrad Wilk of Rage Against the Machine was chosen as the drummer, and Rick Rubin was chosen as the producer.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSICAL_ARTIST, AWARD\n\nText:\nBrad Wilk of Rage Against the Machine was chosen as the drummer, and Rick Rubin was chosen as the producer.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_221", "text": "Many stations play primarily gospel music, including Urban contemporary gospel and Southern Gospel, or contemporary worship music, while others play all formats of contemporary Christian music, including Christian pop, Christian rock, Christian rap, Christian country music, and Christian alternative rock.", "requested_types": ["EVENT", "MUSIC_GENRE"], "distractors": ["EVENT"], "gold": {"EVENT": [], "MUSIC_GENRE": ["gospel music", "Urban contemporary gospel", "Southern Gospel", "contemporary worship music", "contemporary Christian music", "Christian rock", "Christian rap", "Christian country music", "Christian alternative rock"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE\n\nText:\nMany stations play primarily gospel music, including Urban contemporary gospel and Southern Gospel, or contemporary worship music, while others play all formats of contemporary Christian music, including Christian pop, Christian rock, Christian rap, Christian country music, and Christian alternative rock.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE\n\nText:\nMany stations play primarily gospel music, including Urban contemporary gospel and Southern Gospel, or contemporary worship music, while others play all formats of contemporary Christian music, including Christian pop, Christian rock, Christian rap, Christian country music, and Christian alternative rock.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_222", "text": "Williams has received a record eighteen Brit Awards - winning Brit Award for British Male Solo Artist four times, two awards for Outstanding Contribution to Music and the 2017 Brits Icon for his lasting impact on British culture, eight German Echo Music Prize, and three MTV European Music Awards.", "requested_types": ["AWARD", "BAND", "MUSICAL_ARTIST", "MISC"], "distractors": ["BAND"], "gold": {"AWARD": ["Brit Awards", "Brit Award for British Male Solo Artist", "Outstanding Contribution to Music", "2017 Brits Icon", "Echo Music Prize", "MTV European Music Awards"], "BAND": [], "MUSICAL_ARTIST": ["Williams"], "MISC": ["British", "German"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSICAL_ARTIST, MISC\n\nText:\nWilliams has received a record eighteen Brit Awards - winning Brit Award for British Male Solo Artist four times, two awards for Outstanding Contribution to Music and the 2017 Brits Icon for his lasting impact on British culture, eight German Echo Music Prize, and three MTV European Music 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, BAND, MUSICAL_ARTIST, MISC\n\nText:\nWilliams has received a record eighteen Brit Awards - winning Brit Award for British Male Solo Artist four times, two awards for Outstanding Contribution to Music and the 2017 Brits Icon for his lasting impact on British culture, eight German Echo Music Prize, and three MTV European Music 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_music_223", "text": "The album included three songs written or co-written by Dylan (This Wheel's on Fire, Tears of Rage, and I Shall Be Released) as well as The Weight, the use of which in the film Easy Rider would make it one of their best-known songs.", "requested_types": ["MUSICAL_ARTIST", "ORGANISATION", "MISC", "LOCATION", "SONG", "BAND"], "distractors": ["BAND", "ORGANISATION", "LOCATION"], "gold": {"MUSICAL_ARTIST": ["Dylan"], "ORGANISATION": [], "MISC": ["Easy Rider"], "LOCATION": [], "SONG": ["This Wheel's on Fire", "Tears of Rage", "I Shall Be Released", "The Weight"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ORGANISATION, MISC, LOCATION, SONG, BAND\n\nText:\nThe album included three songs written or co-written by Dylan (This Wheel's on Fire, Tears of Rage, and I Shall Be Released) as well as The Weight, the use of which in the film Easy Rider would make it one of their best-known songs.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ORGANISATION, MISC, LOCATION, SONG, BAND\n\nText:\nThe album included three songs written or co-written by Dylan (This Wheel's on Fire, Tears of Rage, and I Shall Be Released) as well as The Weight, the use of which in the film Easy Rider would make it one of their best-known songs.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_224", "text": "Phoenix music venues have included Comerica Theatre, Ashley Furniture HomeStore Pavilion, Long Wong's, The Rebel Lounge (formerly The Mason Jar), Modified Arts, Trunk Space, Paper Heart Gallery, Club Red, Marquee Theatre, The Van Buren, Crescent Ball Room, Valley Bar, the Nile Theater, The Rhythm Room, and Compton Terrace.", "requested_types": ["ORGANISATION", "MUSICAL_ARTIST", "LOCATION"], "distractors": ["ORGANISATION", "MUSICAL_ARTIST"], "gold": {"ORGANISATION": [], "MUSICAL_ARTIST": [], "LOCATION": ["Phoenix", "Comerica Theatre", "Ashley Furniture HomeStore Pavilion", "Long Wong's", "The Rebel Lounge", "The Mason Jar", "Modified Arts", "Trunk Space", "Paper Heart Gallery", "Club Red", "Marquee Theatre", "The Van Buren", "Crescent Ball Room", "Valley Bar", "Nile Theater", "The Rhythm Room", "Compton Terrace"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, LOCATION\n\nText:\nPhoenix music venues have included Comerica Theatre, Ashley Furniture HomeStore Pavilion, Long Wong's, The Rebel Lounge (formerly The Mason Jar), Modified Arts, Trunk Space, Paper Heart Gallery, Club Red, Marquee Theatre, The Van Buren, Crescent Ball Room, Valley Bar, the Nile Theater, The Rhythm Room, and Compton Terrace.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, LOCATION\n\nText:\nPhoenix music venues have included Comerica Theatre, Ashley Furniture HomeStore Pavilion, Long Wong's, The Rebel Lounge (formerly The Mason Jar), Modified Arts, Trunk Space, Paper Heart Gallery, Club Red, Marquee Theatre, The Van Buren, Crescent Ball Room, Valley Bar, the Nile Theater, The Rhythm Room, and Compton Terrace.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_225", "text": "His producing credits also include the UK or US Top 10 albums Cloud Nine (Harrison, 1987), Mystery Girl (Orbison, 1989), Full Moon Fever (Petty, 1989), Into the Great Wide Open (Petty, 1991), Flaming Pie (Paul McCartney, 1997) and Get Up! (Bryan Adams, 2015).", "requested_types": ["BAND", "COUNTRY", "ALBUM", "MUSICAL_ARTIST"], "distractors": ["BAND"], "gold": {"BAND": [], "COUNTRY": ["UK", "US"], "ALBUM": ["Cloud Nine", "Mystery Girl", "Full Moon Fever", "Into the Great Wide Open", "Flaming Pie", "Get Up!"], "MUSICAL_ARTIST": ["Harrison", "Orbison", "Petty", "Paul McCartney", "Bryan Adams"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, COUNTRY, ALBUM, MUSICAL_ARTIST\n\nText:\nHis producing credits also include the UK or US Top 10 albums Cloud Nine (Harrison, 1987), Mystery Girl (Orbison, 1989), Full Moon Fever (Petty, 1989), Into the Great Wide Open (Petty, 1991), Flaming Pie (Paul McCartney, 1997) and Get Up! (Bryan Adams, 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: BAND, COUNTRY, ALBUM, MUSICAL_ARTIST\n\nText:\nHis producing credits also include the UK or US Top 10 albums Cloud Nine (Harrison, 1987), Mystery Girl (Orbison, 1989), Full Moon Fever (Petty, 1989), Into the Great Wide Open (Petty, 1991), Flaming Pie (Paul McCartney, 1997) and Get Up! (Bryan Adams, 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_music_226", "text": "Hawkwind are best known for the song Silver Machine, which became a number three UK hit single in 1972, but they scored further hit singles with Urban Guerrilla (another Top 40 hit) and Shot Down in the Night.", "requested_types": ["COUNTRY", "AWARD", "BAND", "MUSICAL_ARTIST", "SONG"], "distractors": ["AWARD", "MUSICAL_ARTIST"], "gold": {"COUNTRY": ["UK"], "AWARD": [], "BAND": ["Hawkwind"], "MUSICAL_ARTIST": [], "SONG": ["Silver Machine", "Urban Guerrilla", "Shot Down in the Night"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSICAL_ARTIST, SONG\n\nText:\nHawkwind are best known for the song Silver Machine, which became a number three UK hit single in 1972, but they scored further hit singles with Urban Guerrilla (another Top 40 hit) and Shot Down in the Night.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSICAL_ARTIST, SONG\n\nText:\nHawkwind are best known for the song Silver Machine, which became a number three UK hit single in 1972, but they scored further hit singles with Urban Guerrilla (another Top 40 hit) and Shot Down in the Night.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_227", "text": "The latter earned Kidman the Primetime Emmy Award for Primetime Emmy Award for Outstanding Lead Actress in a Limited Series or Movie and Primetime Emmy Award for Outstanding Limited Series.", "requested_types": ["EVENT", "PERSON", "AWARD", "MUSIC_GENRE", "BAND"], "distractors": ["MUSIC_GENRE", "BAND", "EVENT"], "gold": {"EVENT": [], "PERSON": ["Kidman"], "AWARD": ["Primetime Emmy Award", "Primetime Emmy Award for Outstanding Lead Actress in a Limited Series or Movie", "Primetime Emmy Award for Outstanding Limited Series"], "MUSIC_GENRE": [], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, AWARD, MUSIC_GENRE, BAND\n\nText:\nThe latter earned Kidman the Primetime Emmy Award for Primetime Emmy Award for Outstanding Lead Actress in a Limited Series or Movie and Primetime Emmy Award for Outstanding Limited Series.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, AWARD, MUSIC_GENRE, BAND\n\nText:\nThe latter earned Kidman the Primetime Emmy Award for Primetime Emmy Award for Outstanding Lead Actress in a Limited Series or Movie and Primetime Emmy Award for Outstanding Limited Series.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_228", "text": "In 1995 they supported Greenpeace, International Physicians for the Prevention of Nuclear War, Aktion Atomteststop (an initiative for a nuclear test ban), the Bund für Umwelt und Naturschutz Deutschland (German Friends of the Earth chapter) and they were featured on the track Tout Pour Sauver L'Amour (Everything to save love) on the Stop Chirac compilation album.", "requested_types": ["MISC", "SONG", "ORGANISATION", "ALBUM", "MUSIC_GENRE", "BAND", "LOCATION"], "distractors": ["LOCATION", "MUSIC_GENRE", "BAND"], "gold": {"MISC": ["German"], "SONG": ["Tout Pour Sauver L'Amour", "Everything to save love"], "ORGANISATION": ["Greenpeace", "International Physicians for the Prevention of Nuclear War", "Aktion Atomteststop", "Bund für Umwelt und Naturschutz Deutschland", "Friends of the Earth"], "ALBUM": ["Stop Chirac"], "MUSIC_GENRE": [], "BAND": [], "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, SONG, ORGANISATION, ALBUM, MUSIC_GENRE, BAND, LOCATION\n\nText:\nIn 1995 they supported Greenpeace, International Physicians for the Prevention of Nuclear War, Aktion Atomteststop (an initiative for a nuclear test ban), the Bund für Umwelt und Naturschutz Deutschland (German Friends of the Earth chapter) and they were featured on the track Tout Pour Sauver L'Amour (Everything to save love) on the Stop Chirac compilation album.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, ORGANISATION, ALBUM, MUSIC_GENRE, BAND, LOCATION\n\nText:\nIn 1995 they supported Greenpeace, International Physicians for the Prevention of Nuclear War, Aktion Atomteststop (an initiative for a nuclear test ban), the Bund für Umwelt und Naturschutz Deutschland (German Friends of the Earth chapter) and they were featured on the track Tout Pour Sauver L'Amour (Everything to save love) on the Stop Chirac compilation album.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_229", "text": "Martin Fry and band were once more accompanied by the Southbank Sinfonia Orchestra for dates at Liverpool's Philharmonic Hall, Glasgow Royal Concert Hall, Sheffield City Hall, the Theatre Royal, Drury Lane in London and Symphony Hall, Birmingham, between November 4th and 9th, 2015.", "requested_types": ["LOCATION", "BAND", "MUSIC_GENRE", "MUSICAL_ARTIST"], "distractors": ["MUSIC_GENRE"], "gold": {"LOCATION": ["Liverpool", "Philharmonic Hall", "Glasgow Royal Concert Hall", "Sheffield City Hall", "Theatre Royal", "Drury Lane", "London", "Symphony Hall", "Birmingham"], "BAND": ["Southbank Sinfonia Orchestra"], "MUSIC_GENRE": [], "MUSICAL_ARTIST": ["Martin Fry"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nMartin Fry and band were once more accompanied by the Southbank Sinfonia Orchestra for dates at Liverpool's Philharmonic Hall, Glasgow Royal Concert Hall, Sheffield City Hall, the Theatre Royal, Drury Lane in London and Symphony Hall, Birmingham, between November 4th and 9th, 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: LOCATION, BAND, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nMartin Fry and band were once more accompanied by the Southbank Sinfonia Orchestra for dates at Liverpool's Philharmonic Hall, Glasgow Royal Concert Hall, Sheffield City Hall, the Theatre Royal, Drury Lane in London and Symphony Hall, Birmingham, between November 4th and 9th, 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_music_230", "text": "He toured with his band Les Mistigris (not related to Mistigris) in Germany, Belgium, France and Turkey until 1967.", "requested_types": ["ORGANISATION", "MISC", "COUNTRY", "MUSICAL_INSTRUMENT", "AWARD", "BAND"], "distractors": ["AWARD", "MUSICAL_INSTRUMENT", "MISC"], "gold": {"ORGANISATION": ["Mistigris"], "MISC": [], "COUNTRY": ["Germany", "Belgium", "France", "Turkey"], "MUSICAL_INSTRUMENT": [], "AWARD": [], "BAND": ["Les Mistigris"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, AWARD, BAND\n\nText:\nHe toured with his band Les Mistigris (not related to Mistigris) in Germany, Belgium, France and Turkey until 1967.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, AWARD, BAND\n\nText:\nHe toured with his band Les Mistigris (not related to Mistigris) in Germany, Belgium, France and Turkey until 1967.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_231", "text": "The near 3-hour-long epic, which chronicled the saga of the Corleone family, received overwhelmingly positive reviews from critics and fetched Coppola the Academy Awards for Academy Award for Best Adapted Screenplay, which he shared with Mario Puzo and two Golden Globe Award s: for Golden Globe Award for Best Director and Golden Globe Award for Best Screenplay.", "requested_types": ["MISC", "MUSIC_GENRE", "AWARD", "PERSON"], "distractors": ["MUSIC_GENRE"], "gold": {"MISC": ["Corleone family"], "MUSIC_GENRE": [], "AWARD": ["Academy Awards", "Academy Award for Best Adapted Screenplay", "Golden Globe Award", "Golden Globe Award for Best Director", "Golden Globe Award for Best Screenplay"], "PERSON": ["Coppola", "Mario Puzo"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, AWARD, PERSON\n\nText:\nThe near 3-hour-long epic, which chronicled the saga of the Corleone family, received overwhelmingly positive reviews from critics and fetched Coppola the Academy Awards for Academy Award for Best Adapted Screenplay, which he shared with Mario Puzo and two Golden Globe Award s: for Golden Globe Award for Best Director and Golden Globe Award for Best Screenplay.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, AWARD, PERSON\n\nText:\nThe near 3-hour-long epic, which chronicled the saga of the Corleone family, received overwhelmingly positive reviews from critics and fetched Coppola the Academy Awards for Academy Award for Best Adapted Screenplay, which he shared with Mario Puzo and two Golden Globe Award s: for Golden Globe Award for Best Director and Golden Globe Award for Best Screenplay.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_232", "text": "He has been nominated in more categories than any other musician, namely Country music, pop, jazz, Bluegrass music, classical, Folk music, spoken word, composition, and arranging.", "requested_types": ["MUSIC_GENRE", "EVENT", "MISC"], "distractors": ["EVENT"], "gold": {"MUSIC_GENRE": ["Country music", "pop", "jazz", "Bluegrass music", "classical", "Folk music"], "EVENT": [], "MISC": ["spoken word", "composition", "arranging"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, EVENT, MISC\n\nText:\nHe has been nominated in more categories than any other musician, namely Country music, pop, jazz, Bluegrass music, classical, Folk music, spoken word, composition, and arranging.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, EVENT, MISC\n\nText:\nHe has been nominated in more categories than any other musician, namely Country music, pop, jazz, Bluegrass music, classical, Folk music, spoken word, composition, and arranging.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_233", "text": "That January, the San Francisco band Blue Cheer released a cover of Eddie Cochran 's classic Summertime Blues, from their debut album Vincebus Eruptum, that many consider the first TRUE heavy metal recording.McCleary (2004), pp. 240, 506.", "requested_types": ["MUSIC_GENRE", "LOCATION", "MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "ALBUM", "SONG", "EVENT", "BAND"], "distractors": ["MUSICAL_INSTRUMENT", "EVENT"], "gold": {"MUSIC_GENRE": ["heavy metal"], "LOCATION": ["San Francisco"], "MUSICAL_ARTIST": ["Eddie Cochran"], "MUSICAL_INSTRUMENT": [], "ALBUM": ["Vincebus Eruptum"], "SONG": ["Summertime Blues"], "EVENT": [], "BAND": ["Blue Cheer"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, LOCATION, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, ALBUM, SONG, EVENT, BAND\n\nText:\nThat January, the San Francisco band Blue Cheer released a cover of Eddie Cochran 's classic Summertime Blues, from their debut album Vincebus Eruptum, that many consider the first TRUE heavy metal recording.McCleary (2004), pp. 240, 506.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, LOCATION, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, ALBUM, SONG, EVENT, BAND\n\nText:\nThat January, the San Francisco band Blue Cheer released a cover of Eddie Cochran 's classic Summertime Blues, from their debut album Vincebus Eruptum, that many consider the first TRUE heavy metal recording.McCleary (2004), pp. 240, 506.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_234", "text": "It was not long before Peña was touring the world, both as a soloist and an accompanist with performances at Carnegie Hall in New York City, the Royal Albert Hall in London and the Concertgebouw in Amsterdam.", "requested_types": ["EVENT", "LOCATION", "BAND"], "distractors": ["EVENT"], "gold": {"EVENT": [], "LOCATION": ["Carnegie Hall", "New York City", "Royal Albert Hall", "London", "Concertgebouw", "Amsterdam"], "BAND": ["Peña"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nIt was not long before Peña was touring the world, both as a soloist and an accompanist with performances at Carnegie Hall in New York City, the Royal Albert Hall in London and the Concertgebouw in Amsterdam.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nIt was not long before Peña was touring the world, both as a soloist and an accompanist with performances at Carnegie Hall in New York City, the Royal Albert Hall in London and the Concertgebouw in Amsterdam.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_235", "text": "It includes folklore music of parts of Bolivia, Ecuador, Chile, Colombia, Peru and Venezuela.", "requested_types": ["MUSICAL_ARTIST", "ORGANISATION", "COUNTRY", "MUSIC_GENRE"], "distractors": ["ORGANISATION", "MUSICAL_ARTIST"], "gold": {"MUSICAL_ARTIST": [], "ORGANISATION": [], "COUNTRY": ["Bolivia", "Ecuador", "Chile", "Colombia", "Peru", "Venezuela"], "MUSIC_GENRE": ["folklore music"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ORGANISATION, COUNTRY, MUSIC_GENRE\n\nText:\nIt includes folklore music of parts of Bolivia, Ecuador, Chile, Colombia, Peru and Venezuela.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ORGANISATION, COUNTRY, MUSIC_GENRE\n\nText:\nIt includes folklore music of parts of Bolivia, Ecuador, Chile, Colombia, Peru and Venezuela.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_236", "text": "Collette supports various charities including Médecins Sans Frontières, Amnesty International and Feeding America.", "requested_types": ["MISC", "MUSICAL_ARTIST", "LOCATION", "ORGANISATION"], "distractors": ["LOCATION", "MISC"], "gold": {"MISC": [], "MUSICAL_ARTIST": ["Collette"], "LOCATION": [], "ORGANISATION": ["Médecins Sans Frontières", "Amnesty International", "Feeding America"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, LOCATION, ORGANISATION\n\nText:\nCollette supports various charities including Médecins Sans Frontières, Amnesty International and Feeding 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: MISC, MUSICAL_ARTIST, LOCATION, ORGANISATION\n\nText:\nCollette supports various charities including Médecins Sans Frontières, Amnesty International and Feeding 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_music_237", "text": "After several line-up changes, the band went on to release a series of UK and US platinum and gold albums, including 1982's The Number of the Beast, 1983's Piece of Mind, 1984's Powerslave, 1985's live release Live After Death, 1986's Somewhere in Time, and 1988's Seventh Son of a Seventh Son.", "requested_types": ["EVENT", "LOCATION", "COUNTRY", "ALBUM", "SONG"], "distractors": ["SONG", "EVENT", "LOCATION"], "gold": {"EVENT": [], "LOCATION": [], "COUNTRY": ["UK", "US"], "ALBUM": ["The Number of the Beast", "Piece of Mind", "Powerslave", "Live After Death", "Somewhere in Time", "Seventh Son of a Seventh Son"], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, SONG\n\nText:\nAfter several line-up changes, the band went on to release a series of UK and US platinum and gold albums, including 1982's The Number of the Beast, 1983's Piece of Mind, 1984's Powerslave, 1985's live release Live After Death, 1986's Somewhere in Time, and 1988's Seventh Son of a Seventh Son.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, SONG\n\nText:\nAfter several line-up changes, the band went on to release a series of UK and US platinum and gold albums, including 1982's The Number of the Beast, 1983's Piece of Mind, 1984's Powerslave, 1985's live release Live After Death, 1986's Somewhere in Time, and 1988's Seventh Son of a Seventh Son.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_238", "text": "In 2014, the Robert David MacDonald and Jeremy Sams translation (previously used in 1994 at the Donmar Warehouse) toured the UK, presented by the Graeae Theatre Company with Nottingham Playhouse, New Wolsey Theatre Ipswich, Birmingham Repertory Theatre and West Yorkshire Playhouse.", "requested_types": ["MISC", "ORGANISATION", "PERSON", "COUNTRY", "LOCATION"], "distractors": ["MISC"], "gold": {"MISC": [], "ORGANISATION": ["Graeae Theatre Company", "Nottingham Playhouse", "New Wolsey Theatre", "Birmingham Repertory Theatre", "West Yorkshire Playhouse"], "PERSON": ["Robert David MacDonald", "Jeremy Sams"], "COUNTRY": ["UK"], "LOCATION": ["Donmar Warehouse"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, COUNTRY, LOCATION\n\nText:\nIn 2014, the Robert David MacDonald and Jeremy Sams translation (previously used in 1994 at the Donmar Warehouse) toured the UK, presented by the Graeae Theatre Company with Nottingham Playhouse, New Wolsey Theatre Ipswich, Birmingham Repertory Theatre and West Yorkshire Playhouse.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, COUNTRY, LOCATION\n\nText:\nIn 2014, the Robert David MacDonald and Jeremy Sams translation (previously used in 1994 at the Donmar Warehouse) toured the UK, presented by the Graeae Theatre Company with Nottingham Playhouse, New Wolsey Theatre Ipswich, Birmingham Repertory Theatre and West Yorkshire Playhouse.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_239", "text": "Rob Sheffield in an AllMusic review feels that, like Mellow Gold, Odelay incorporates elements from various genres, including Folk music and Country music, grungy garage rock, stiff-boned electro, louche exotica, Old-school hip hop and noise rock.", "requested_types": ["MUSIC_GENRE", "ALBUM", "MUSICAL_ARTIST", "ORGANISATION", "SONG", "BAND"], "distractors": ["BAND", "SONG"], "gold": {"MUSIC_GENRE": ["Folk music", "Country music", "grungy garage rock", "electro", "exotica", "Old-school hip hop", "noise rock"], "ALBUM": ["Mellow Gold", "Odelay"], "MUSICAL_ARTIST": ["Rob Sheffield"], "ORGANISATION": ["AllMusic"], "SONG": [], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, ALBUM, MUSICAL_ARTIST, ORGANISATION, SONG, BAND\n\nText:\nRob Sheffield in an AllMusic review feels that, like Mellow Gold, Odelay incorporates elements from various genres, including Folk music and Country music, grungy garage rock, stiff-boned electro, louche exotica, Old-school hip hop and noise rock.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, ALBUM, MUSICAL_ARTIST, ORGANISATION, SONG, BAND\n\nText:\nRob Sheffield in an AllMusic review feels that, like Mellow Gold, Odelay incorporates elements from various genres, including Folk music and Country music, grungy garage rock, stiff-boned electro, louche exotica, Old-school hip hop and noise rock.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_240", "text": "Apple's debut album earned her a Grammy Award for Grammy Award for Best Female Rock Vocal Performance for Criminal and the MTV Video Music Award for MTV Video Music Award for Best New Artist for Sleep to Dream.", "requested_types": ["COUNTRY", "MUSIC_GENRE", "SONG", "ALBUM", "MUSICAL_ARTIST", "AWARD"], "distractors": ["MUSIC_GENRE", "ALBUM", "COUNTRY"], "gold": {"COUNTRY": [], "MUSIC_GENRE": [], "SONG": ["Criminal", "Sleep to Dream"], "ALBUM": [], "MUSICAL_ARTIST": ["Apple"], "AWARD": ["Grammy Award", "Grammy Award for Best Female Rock Vocal Performance", "MTV Video Music Award", "MTV Video Music Award for Best New Artist"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, SONG, ALBUM, MUSICAL_ARTIST, AWARD\n\nText:\nApple's debut album earned her a Grammy Award for Grammy Award for Best Female Rock Vocal Performance for Criminal and the MTV Video Music Award for MTV Video Music Award for Best New Artist for Sleep to Dream.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, SONG, ALBUM, MUSICAL_ARTIST, AWARD\n\nText:\nApple's debut album earned her a Grammy Award for Grammy Award for Best Female Rock Vocal Performance for Criminal and the MTV Video Music Award for MTV Video Music Award for Best New Artist for Sleep to Dream.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_241", "text": "The focus of their activities in Europe became annual concert cycles at the Wiener Konzerthaus, at London's Queen Elizabeth Hall, the Frankfurt Alte Oper, the Théâtre des Champs Elysées in Paris, the Philharmonic Hall in Cologne, the Zurich Opera, as well as regular concerts at most major halls and venues around the world (among them La Scala, Concertgebouw, Berliner Philharmonie, Carnegie Hall, Teatro Colón, Suntory Hall, etc.", "requested_types": ["MISC", "PERSON", "LOCATION"], "distractors": ["PERSON", "MISC"], "gold": {"MISC": [], "PERSON": [], "LOCATION": ["Europe", "Wiener Konzerthaus", "London", "Queen Elizabeth Hall", "Frankfurt Alte Oper", "Théâtre des Champs Elysées", "Paris", "Philharmonic Hall", "Cologne", "Zurich Opera", "La Scala", "Concertgebouw", "Berliner Philharmonie", "Carnegie Hall", "Teatro Colón", "Suntory Hall"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION\n\nText:\nThe focus of their activities in Europe became annual concert cycles at the Wiener Konzerthaus, at London's Queen Elizabeth Hall, the Frankfurt Alte Oper, the Théâtre des Champs Elysées in Paris, the Philharmonic Hall in Cologne, the Zurich Opera, as well as regular concerts at most major halls and venues around the world (among them La Scala, Concertgebouw, Berliner Philharmonie, Carnegie Hall, Teatro Colón, Suntory Hall, 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: MISC, PERSON, LOCATION\n\nText:\nThe focus of their activities in Europe became annual concert cycles at the Wiener Konzerthaus, at London's Queen Elizabeth Hall, the Frankfurt Alte Oper, the Théâtre des Champs Elysées in Paris, the Philharmonic Hall in Cologne, the Zurich Opera, as well as regular concerts at most major halls and venues around the world (among them La Scala, Concertgebouw, Berliner Philharmonie, Carnegie Hall, Teatro Colón, Suntory Hall, 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_music_242", "text": "The original 2005 Broadway theatre production was nominated for 14 Tony Awards and won three: Tony Award for Best Musical, Tony Award for Best Featured Actress in a Musical (Sara Ramirez), and Tony Award for Best Direction of a Musical (Mike Nichols).", "requested_types": ["ORGANISATION", "MUSICAL_ARTIST", "AWARD", "EVENT"], "distractors": ["EVENT"], "gold": {"ORGANISATION": ["Broadway theatre"], "MUSICAL_ARTIST": ["Sara Ramirez", "Mike Nichols"], "AWARD": ["Tony Awards", "Tony Award for Best Musical", "Tony Award for Best Featured Actress in a Musical", "Tony Award for Best Direction of a Musical"], "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, MUSICAL_ARTIST, AWARD, EVENT\n\nText:\nThe original 2005 Broadway theatre production was nominated for 14 Tony Awards and won three: Tony Award for Best Musical, Tony Award for Best Featured Actress in a Musical (Sara Ramirez), and Tony Award for Best Direction of a Musical (Mike Nichols).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, AWARD, EVENT\n\nText:\nThe original 2005 Broadway theatre production was nominated for 14 Tony Awards and won three: Tony Award for Best Musical, Tony Award for Best Featured Actress in a Musical (Sara Ramirez), and Tony Award for Best Direction of a Musical (Mike Nichols).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_243", "text": "Country music, also known as country and western (or simply country), and hillbilly music, is a genre of popular music that takes its roots from genres such as blues and old-time music, and various types of American folk music including Appalachian music, Cajun music, and the cowboy Western music styles of Red Dirt, New Mexico music, Texas country music, and Tejano music.", "requested_types": ["LOCATION", "MISC", "MUSIC_GENRE", "BAND", "AWARD"], "distractors": ["BAND", "AWARD", "LOCATION"], "gold": {"LOCATION": [], "MISC": ["Appalachian music", "Cajun music", "cowboy Western music"], "MUSIC_GENRE": ["Country music", "country", "western", "hillbilly music", "popular music", "blues", "old-time music", "American folk music", "Red Dirt", "New Mexico music", "Texas country music", "Tejano music"], "BAND": [], "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: LOCATION, MISC, MUSIC_GENRE, BAND, AWARD\n\nText:\nCountry music, also known as country and western (or simply country), and hillbilly music, is a genre of popular music that takes its roots from genres such as blues and old-time music, and various types of American folk music including Appalachian music, Cajun music, and the cowboy Western music styles of Red Dirt, New Mexico music, Texas country music, and Tejano music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, BAND, AWARD\n\nText:\nCountry music, also known as country and western (or simply country), and hillbilly music, is a genre of popular music that takes its roots from genres such as blues and old-time music, and various types of American folk music including Appalachian music, Cajun music, and the cowboy Western music styles of Red Dirt, New Mexico music, Texas country music, and Tejano music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_244", "text": "He also produced tracks for a number of other acts on Ruthless Records, including Eazy-E's 1988 solo debut Eazy-Duz-It, Above the Law 's 1990 debut Livin' Like Hustlers, Michel'le 's 1989 self-titled debut, The D.O.C.' s 1989 debut No One Can Do It Better, J.J. Fad 's 1988 debut Supersonic and funk rock musician Jimmy Z' s 1991 album Muzical Madness", "requested_types": ["ALBUM", "MUSICAL_ARTIST", "ORGANISATION", "MUSIC_GENRE", "SONG", "BAND", "PERSON"], "distractors": ["PERSON", "SONG"], "gold": {"ALBUM": ["Eazy-Duz-It", "Livin 'Like Hustlers", "No One Can Do It Better", "Supersonic", "Muzical Madness"], "MUSICAL_ARTIST": ["Eazy-E", "Michel'le", "The D.O.C.", "Jimmy Z"], "ORGANISATION": ["Ruthless Records"], "MUSIC_GENRE": ["funk rock"], "SONG": [], "BAND": ["Above the Law", "J.J. Fad"], "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: ALBUM, MUSICAL_ARTIST, ORGANISATION, MUSIC_GENRE, SONG, BAND, PERSON\n\nText:\nHe also produced tracks for a number of other acts on Ruthless Records, including Eazy-E's 1988 solo debut Eazy-Duz-It, Above the Law 's 1990 debut Livin' Like Hustlers, Michel'le 's 1989 self-titled debut, The D.O.C.' s 1989 debut No One Can Do It Better, J.J. Fad 's 1988 debut Supersonic and funk rock musician Jimmy Z' s 1991 album Muzical Madness\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_ARTIST, ORGANISATION, MUSIC_GENRE, SONG, BAND, PERSON\n\nText:\nHe also produced tracks for a number of other acts on Ruthless Records, including Eazy-E's 1988 solo debut Eazy-Duz-It, Above the Law 's 1990 debut Livin' Like Hustlers, Michel'le 's 1989 self-titled debut, The D.O.C.' s 1989 debut No One Can Do It Better, J.J. Fad 's 1988 debut Supersonic and funk rock musician Jimmy Z' s 1991 album Muzical Madness\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_245", "text": "In 2004, Grohl drummed on several tracks for Nine Inch Nails' 2005 album With Teeth, later returning to play drums on 'The Idea of You' from their 2016 EP Not the Actual Events.", "requested_types": ["BAND", "MUSIC_GENRE", "MUSICAL_INSTRUMENT", "MUSICAL_ARTIST", "SONG", "ALBUM"], "distractors": ["MUSIC_GENRE"], "gold": {"BAND": ["Nine Inch Nails"], "MUSIC_GENRE": [], "MUSICAL_INSTRUMENT": ["drums"], "MUSICAL_ARTIST": ["Grohl"], "SONG": ["The Idea of You"], "ALBUM": ["With Teeth", "Not the Actual Events"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, SONG, ALBUM\n\nText:\nIn 2004, Grohl drummed on several tracks for Nine Inch Nails' 2005 album With Teeth, later returning to play drums on 'The Idea of You' from their 2016 EP Not the Actual Events.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, SONG, ALBUM\n\nText:\nIn 2004, Grohl drummed on several tracks for Nine Inch Nails' 2005 album With Teeth, later returning to play drums on 'The Idea of You' from their 2016 EP Not the Actual Events.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_246", "text": "Love has been candid about her diverse musical influences, the earliest being Patti Smith, The Runaways, and The Pretenders, artists she discovered while in juvenile hall at age fifteen.", "requested_types": ["BAND", "PERSON", "ORGANISATION", "MUSICAL_ARTIST"], "distractors": ["ORGANISATION", "PERSON"], "gold": {"BAND": ["The Runaways", "The Pretenders"], "PERSON": [], "ORGANISATION": [], "MUSICAL_ARTIST": ["Love", "Patti Smith"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, PERSON, ORGANISATION, MUSICAL_ARTIST\n\nText:\nLove has been candid about her diverse musical influences, the earliest being Patti Smith, The Runaways, and The Pretenders, artists she discovered while in juvenile hall at age fifteen.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, PERSON, ORGANISATION, MUSICAL_ARTIST\n\nText:\nLove has been candid about her diverse musical influences, the earliest being Patti Smith, The Runaways, and The Pretenders, artists she discovered while in juvenile hall at age fifteen.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_247", "text": "In 1999, it was certified 3 × platinum by the British Phonographic Industry (BPI), 3 × Platinum by the Australian Record Industry Association and platinum by the Recording Industry Association of America (RIAA).", "requested_types": ["BAND", "MISC", "ORGANISATION"], "distractors": ["MISC", "BAND"], "gold": {"BAND": [], "MISC": [], "ORGANISATION": ["British Phonographic Industry", "BPI", "Australian Record Industry Association", "Recording Industry Association of America", "RIAA"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MISC, ORGANISATION\n\nText:\nIn 1999, it was certified 3 × platinum by the British Phonographic Industry (BPI), 3 × Platinum by the Australian Record Industry Association and platinum by the Recording Industry Association of America (RIAA).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MISC, ORGANISATION\n\nText:\nIn 1999, it was certified 3 × platinum by the British Phonographic Industry (BPI), 3 × Platinum by the Australian Record Industry Association and platinum by the Recording Industry Association of America (RIAA).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_248", "text": "Every style of music was offered, from Rock music and Pop music to Spanish-language programming (for Mexican restaurants), jazz, blues, classical and even easy listening.", "requested_types": ["MISC", "MUSIC_GENRE", "ALBUM", "MUSICAL_ARTIST"], "distractors": ["ALBUM", "MUSICAL_ARTIST"], "gold": {"MISC": ["Spanish-language", "Mexican"], "MUSIC_GENRE": ["Rock music", "Pop music", "jazz", "blues", "classical", "easy listening"], "ALBUM": [], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, ALBUM, MUSICAL_ARTIST\n\nText:\nEvery style of music was offered, from Rock music and Pop music to Spanish-language programming (for Mexican restaurants), jazz, blues, classical and even easy listening.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, ALBUM, MUSICAL_ARTIST\n\nText:\nEvery style of music was offered, from Rock music and Pop music to Spanish-language programming (for Mexican restaurants), jazz, blues, classical and even easy listening.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_249", "text": "Singers such as Blind Willie McTell and Blind Boy Fuller performed in the southeastern delicate and lyrical Piedmont blues tradition, which used an elaborate ragtime-based fingerpicking guitar technique.", "requested_types": ["MUSICAL_ARTIST", "AWARD", "MISC", "BAND", "MUSIC_GENRE"], "distractors": ["AWARD", "BAND"], "gold": {"MUSICAL_ARTIST": ["Blind Willie McTell", "Blind Boy Fuller"], "AWARD": [], "MISC": ["ragtime-based", "fingerpicking guitar technique"], "BAND": [], "MUSIC_GENRE": ["Piedmont blues"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, AWARD, MISC, BAND, MUSIC_GENRE\n\nText:\nSingers such as Blind Willie McTell and Blind Boy Fuller performed in the southeastern delicate and lyrical Piedmont blues tradition, which used an elaborate ragtime-based fingerpicking guitar technique.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, AWARD, MISC, BAND, MUSIC_GENRE\n\nText:\nSingers such as Blind Willie McTell and Blind Boy Fuller performed in the southeastern delicate and lyrical Piedmont blues tradition, which used an elaborate ragtime-based fingerpicking guitar technique.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_250", "text": "The band also saw success with the rest of the album's singles, Bleed It Out, Shadow of the Day, Given Up, and Leave Out All the Rest, which were released throughout 2007 and early 2008.", "requested_types": ["EVENT", "ORGANISATION", "SONG", "MUSIC_GENRE"], "distractors": ["ORGANISATION", "EVENT", "MUSIC_GENRE"], "gold": {"EVENT": [], "ORGANISATION": [], "SONG": ["Bleed It Out", "Shadow of the Day", "Given Up", "Leave Out All the Rest"], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE\n\nText:\nThe band also saw success with the rest of the album's singles, Bleed It Out, Shadow of the Day, Given Up, and Leave Out All the Rest, which were released throughout 2007 and early 2008.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE\n\nText:\nThe band also saw success with the rest of the album's singles, Bleed It Out, Shadow of the Day, Given Up, and Leave Out All the Rest, which were released throughout 2007 and early 2008.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_251", "text": "In 1970, Korner and Thorup formed a big-band ensemble, CCS - short for The Collective Consciousness Society - which had several hit singles produced by Mickie Most, including a version of Led Zeppelin 's Whole Lotta Love, which was used as the theme for BBC's Top of the Pops between 1970 and 1981.", "requested_types": ["ORGANISATION", "SONG", "MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "MISC", "LOCATION", "BAND"], "distractors": ["MUSICAL_INSTRUMENT", "LOCATION"], "gold": {"ORGANISATION": ["BBC"], "SONG": ["Whole Lotta Love"], "MUSICAL_ARTIST": ["Korner", "Thorup", "Mickie Most"], "MUSICAL_INSTRUMENT": [], "MISC": ["Top of the Pops"], "LOCATION": [], "BAND": ["CCS", "Collective Consciousness Society", "Led Zeppelin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, MISC, LOCATION, BAND\n\nText:\nIn 1970, Korner and Thorup formed a big-band ensemble, CCS - short for The Collective Consciousness Society - which had several hit singles produced by Mickie Most, including a version of Led Zeppelin 's Whole Lotta Love, which was used as the theme for BBC's Top of the Pops between 1970 and 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: ORGANISATION, SONG, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, MISC, LOCATION, BAND\n\nText:\nIn 1970, Korner and Thorup formed a big-band ensemble, CCS - short for The Collective Consciousness Society - which had several hit singles produced by Mickie Most, including a version of Led Zeppelin 's Whole Lotta Love, which was used as the theme for BBC's Top of the Pops between 1970 and 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_music_252", "text": "Reed's 1984 album New Sensations marked the first time that Reed had charted within the US Top 100 since 1978's Street Hassle, and the first time that Reed had charted in the UK altogether since 1976's Coney Island Baby.", "requested_types": ["COUNTRY", "ALBUM", "PERSON", "MUSICAL_ARTIST"], "distractors": ["PERSON"], "gold": {"COUNTRY": ["US", "UK"], "ALBUM": ["New Sensations", "Street Hassle", "Coney Island Baby"], "PERSON": [], "MUSICAL_ARTIST": ["Reed"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, PERSON, MUSICAL_ARTIST\n\nText:\nReed's 1984 album New Sensations marked the first time that Reed had charted within the US Top 100 since 1978's Street Hassle, and the first time that Reed had charted in the UK altogether since 1976's Coney Island Baby.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, PERSON, MUSICAL_ARTIST\n\nText:\nReed's 1984 album New Sensations marked the first time that Reed had charted within the US Top 100 since 1978's Street Hassle, and the first time that Reed had charted in the UK altogether since 1976's Coney Island Baby.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_253", "text": "Franklin continued to record acclaimed albums such as I Never Loved a Man the Way I Love You (1967), Lady Soul (1968), Spirit in the Dark (1970), Young, Gifted and Black (1972), Amazing Grace (1972), and Sparkle (1976) before experiencing problems with her record company.", "requested_types": ["MUSICAL_INSTRUMENT", "MUSICAL_ARTIST", "ALBUM", "MISC", "SONG"], "distractors": ["MISC", "SONG", "MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_INSTRUMENT": [], "MUSICAL_ARTIST": ["Franklin"], "ALBUM": ["I Never Loved a Man the Way I Love You", "Lady Soul", "Spirit in the Dark", "Young, Gifted and Black", "Amazing Grace", "Sparkle"], "MISC": [], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSICAL_ARTIST, ALBUM, MISC, SONG\n\nText:\nFranklin continued to record acclaimed albums such as I Never Loved a Man the Way I Love You (1967), Lady Soul (1968), Spirit in the Dark (1970), Young, Gifted and Black (1972), Amazing Grace (1972), and Sparkle (1976) before experiencing problems with her record company.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSICAL_ARTIST, ALBUM, MISC, SONG\n\nText:\nFranklin continued to record acclaimed albums such as I Never Loved a Man the Way I Love You (1967), Lady Soul (1968), Spirit in the Dark (1970), Young, Gifted and Black (1972), Amazing Grace (1972), and Sparkle (1976) before experiencing problems with her record company.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_254", "text": "The movie also featured a soundtrack of popular songs, including a cover version of The Troggs' Love Is All Around performed by Wet Wet Wet that remained at number 1 on the UK Singles Chart for fifteen weeks and was then the ninth (now twelfth) biggest selling single of all time in Britain.", "requested_types": ["LOCATION", "BAND", "MISC", "COUNTRY", "SONG", "PERSON"], "distractors": ["PERSON", "LOCATION"], "gold": {"LOCATION": [], "BAND": ["The Troggs", "Wet Wet Wet"], "MISC": ["UK Singles Chart"], "COUNTRY": ["Britain"], "SONG": ["Love Is All Around"], "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, BAND, MISC, COUNTRY, SONG, PERSON\n\nText:\nThe movie also featured a soundtrack of popular songs, including a cover version of The Troggs' Love Is All Around performed by Wet Wet Wet that remained at number 1 on the UK Singles Chart for fifteen weeks and was then the ninth (now twelfth) biggest selling single of all time in Britain.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MISC, COUNTRY, SONG, PERSON\n\nText:\nThe movie also featured a soundtrack of popular songs, including a cover version of The Troggs' Love Is All Around performed by Wet Wet Wet that remained at number 1 on the UK Singles Chart for fifteen weeks and was then the ninth (now twelfth) biggest selling single of all time in Britain.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_255", "text": "Benson also did a version of The Beatles' s 1969 album Abbey Road called The Other Side of Abbey Road, also released in 1969, and a version of White Rabbit, originally written and recorded by San Francisco rock group Great Society, and made famous by Jefferson Airplane.", "requested_types": ["LOCATION", "EVENT", "PERSON", "ALBUM", "BAND", "MUSIC_GENRE", "MUSICAL_INSTRUMENT", "SONG", "MISC"], "distractors": ["MUSICAL_INSTRUMENT", "MISC", "EVENT"], "gold": {"LOCATION": ["San Francisco"], "EVENT": [], "PERSON": ["Benson"], "ALBUM": ["Abbey Road", "The Other Side of Abbey Road"], "BAND": ["The Beatles", "Great Society", "Jefferson Airplane"], "MUSIC_GENRE": ["rock"], "MUSICAL_INSTRUMENT": [], "SONG": ["White Rabbit"], "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: LOCATION, EVENT, PERSON, ALBUM, BAND, MUSIC_GENRE, MUSICAL_INSTRUMENT, SONG, MISC\n\nText:\nBenson also did a version of The Beatles' s 1969 album Abbey Road called The Other Side of Abbey Road, also released in 1969, and a version of White Rabbit, originally written and recorded by San Francisco rock group Great Society, and made famous by Jefferson Airplane.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, ALBUM, BAND, MUSIC_GENRE, MUSICAL_INSTRUMENT, SONG, MISC\n\nText:\nBenson also did a version of The Beatles' s 1969 album Abbey Road called The Other Side of Abbey Road, also released in 1969, and a version of White Rabbit, originally written and recorded by San Francisco rock group Great Society, and made famous by Jefferson Airplane.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_256", "text": "During the course of her career, Saariaho has received commissions from the Lincoln Center for the Kronos Quartet and from IRCAM for the Ensemble Intercontemporain, the BBC, the New York Philharmonic, the Salzburg Music Festival, the Théâtre du Châtelet in Paris, and the Finnish National Opera, among others.", "requested_types": ["MUSICAL_ARTIST", "ORGANISATION", "COUNTRY", "EVENT", "LOCATION", "BAND"], "distractors": ["COUNTRY"], "gold": {"MUSICAL_ARTIST": ["Saariaho"], "ORGANISATION": ["Lincoln Center", "IRCAM", "BBC"], "COUNTRY": [], "EVENT": ["Salzburg Music Festival"], "LOCATION": ["Théâtre du Châtelet", "Paris", "Finnish National Opera"], "BAND": ["Kronos Quartet", "Ensemble Intercontemporain", "New York Philharmonic"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, ORGANISATION, COUNTRY, EVENT, LOCATION, BAND\n\nText:\nDuring the course of her career, Saariaho has received commissions from the Lincoln Center for the Kronos Quartet and from IRCAM for the Ensemble Intercontemporain, the BBC, the New York Philharmonic, the Salzburg Music Festival, the Théâtre du Châtelet in Paris, and the Finnish National Opera, 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: MUSICAL_ARTIST, ORGANISATION, COUNTRY, EVENT, LOCATION, BAND\n\nText:\nDuring the course of her career, Saariaho has received commissions from the Lincoln Center for the Kronos Quartet and from IRCAM for the Ensemble Intercontemporain, the BBC, the New York Philharmonic, the Salzburg Music Festival, the Théâtre du Châtelet in Paris, and the Finnish National Opera, 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_music_257", "text": "The first known opera from Turkey (the Ottoman Empire) was Arshak II, which was an Armenia n opera composed by an ethnic Armenian composer Tigran Chukhajian in 1868 and partially performed in 1873.", "requested_types": ["MISC", "PERSON", "BAND", "COUNTRY"], "distractors": ["BAND"], "gold": {"MISC": ["Arshak II", "Armenian"], "PERSON": ["Tigran Chukhajian"], "BAND": [], "COUNTRY": ["Turkey", "Ottoman Empire", "Armenia"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, COUNTRY\n\nText:\nThe first known opera from Turkey (the Ottoman Empire) was Arshak II, which was an Armenia n opera composed by an ethnic Armenian composer Tigran Chukhajian in 1868 and partially performed in 1873.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, COUNTRY\n\nText:\nThe first known opera from Turkey (the Ottoman Empire) was Arshak II, which was an Armenia n opera composed by an ethnic Armenian composer Tigran Chukhajian in 1868 and partially performed in 1873.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_258", "text": "Hank Williams, Jr (and, to an even greater extent, Hank Williams III), Gary Allan, Shania Twain, Brooks & amp; Dunn, Faith Hill, Garth Brooks, Alan Jackson, Dwight Yoakam, Steve Earle, Dolly Parton, Rosanne Cash and Linda Ronstadt moved country further towards rock influence.", "requested_types": ["MUSICAL_ARTIST", "MUSIC_GENRE", "SONG", "BAND"], "distractors": ["SONG"], "gold": {"MUSICAL_ARTIST": ["Hank Williams, Jr", "Hank Williams III", "Gary Allan", "Shania Twain", "Faith Hill", "Garth Brooks", "Alan Jackson", "Dwight Yoakam", "Steve Earle", "Dolly Parton", "Rosanne Cash", "Linda Ronstadt"], "MUSIC_GENRE": ["rock"], "SONG": [], "BAND": ["Brooks & amp; Dunn"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MUSIC_GENRE, SONG, BAND\n\nText:\nHank Williams, Jr (and, to an even greater extent, Hank Williams III), Gary Allan, Shania Twain, Brooks & amp; Dunn, Faith Hill, Garth Brooks, Alan Jackson, Dwight Yoakam, Steve Earle, Dolly Parton, Rosanne Cash and Linda Ronstadt moved country further towards rock influence.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MUSIC_GENRE, SONG, BAND\n\nText:\nHank Williams, Jr (and, to an even greater extent, Hank Williams III), Gary Allan, Shania Twain, Brooks & amp; Dunn, Faith Hill, Garth Brooks, Alan Jackson, Dwight Yoakam, Steve Earle, Dolly Parton, Rosanne Cash and Linda Ronstadt moved country further towards rock influence.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_259", "text": "On numerous songs from band's fourth album, A Thousand Suns, such as the album's singles (The Catalyst, Burning in the Skies, Iridescent), both Shinoda and Bennington sing.", "requested_types": ["COUNTRY", "MUSICAL_ARTIST", "SONG", "EVENT", "ALBUM"], "distractors": ["COUNTRY", "EVENT"], "gold": {"COUNTRY": [], "MUSICAL_ARTIST": ["Shinoda", "Bennington"], "SONG": ["The Catalyst", "Burning in the Skies", "Iridescent"], "EVENT": [], "ALBUM": ["A Thousand Suns"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, SONG, EVENT, ALBUM\n\nText:\nOn numerous songs from band's fourth album, A Thousand Suns, such as the album's singles (The Catalyst, Burning in the Skies, Iridescent), both Shinoda and Bennington sing.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, SONG, EVENT, ALBUM\n\nText:\nOn numerous songs from band's fourth album, A Thousand Suns, such as the album's singles (The Catalyst, Burning in the Skies, Iridescent), both Shinoda and Bennington sing.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_260", "text": "The stadium hosted the 2003 World Championships in Athletics and from 1999 to 2016 it hosted the annual Meeting Areva athletics meet.", "requested_types": ["EVENT", "SONG", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_ARTIST", "SONG"], "gold": {"EVENT": ["2003 World Championships in Athletics", "Meeting Areva"], "SONG": [], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST\n\nText:\nThe stadium hosted the 2003 World Championships in Athletics and from 1999 to 2016 it hosted the annual Meeting Areva athletics meet.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST\n\nText:\nThe stadium hosted the 2003 World Championships in Athletics and from 1999 to 2016 it hosted the annual Meeting Areva athletics meet.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_261", "text": "Since the band's 1960s heyday, the influence of the Byrds on successive generations of Rock music and Pop music musicians has grown steadily, with acts such as the Eagles, Big Star, Tom Petty & amp; the Heartbreakers, R.E.M., the Bangles, the Smiths, and innumerable alternative rock bands of the post- Punk rock era all exhibiting signs of their influence.", "requested_types": ["SONG", "MUSIC_GENRE", "MISC", "BAND"], "distractors": ["SONG"], "gold": {"SONG": [], "MUSIC_GENRE": ["Rock music", "Punk rock"], "MISC": ["Pop music musicians", "alternative rock bands"], "BAND": ["Byrds", "Eagles", "Big Star", "Tom Petty & amp; the Heartbreakers", "R.E.M.", "the Bangles", "the Smiths"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE, MISC, BAND\n\nText:\nSince the band's 1960s heyday, the influence of the Byrds on successive generations of Rock music and Pop music musicians has grown steadily, with acts such as the Eagles, Big Star, Tom Petty & amp; the Heartbreakers, R.E.M., the Bangles, the Smiths, and innumerable alternative rock bands of the post- Punk rock era all exhibiting signs of their influence.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE, MISC, BAND\n\nText:\nSince the band's 1960s heyday, the influence of the Byrds on successive generations of Rock music and Pop music musicians has grown steadily, with acts such as the Eagles, Big Star, Tom Petty & amp; the Heartbreakers, R.E.M., the Bangles, the Smiths, and innumerable alternative rock bands of the post- Punk rock era all exhibiting signs of their influence.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_262", "text": "The band headlined the Soul Assassins tour with House of Pain and Funkdoobiest as support, then performed on a college tour with Rage Against the Machine and Seven Year Bitch.", "requested_types": ["MUSICAL_INSTRUMENT", "BAND", "MISC"], "distractors": ["MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_INSTRUMENT": [], "BAND": ["House of Pain", "Funkdoobiest", "Rage Against the Machine", "Seven Year Bitch"], "MISC": ["Soul Assassins tour"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, BAND, MISC\n\nText:\nThe band headlined the Soul Assassins tour with House of Pain and Funkdoobiest as support, then performed on a college tour with Rage Against the Machine and Seven Year Bitch.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, BAND, MISC\n\nText:\nThe band headlined the Soul Assassins tour with House of Pain and Funkdoobiest as support, then performed on a college tour with Rage Against the Machine and Seven Year Bitch.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_263", "text": "The Elephant Man was a huge critical and commercial success, and earned eight Academy Awards nominations, including Academy Award for Best Director and Academy Award for Best Adapted Screenplay for Lynch personally.", "requested_types": ["MISC", "BAND", "AWARD", "MUSICAL_ARTIST"], "distractors": ["BAND"], "gold": {"MISC": ["The Elephant Man"], "BAND": [], "AWARD": ["Academy Awards", "Academy Award for Best Director", "Academy Award for Best Adapted Screenplay"], "MUSICAL_ARTIST": ["Lynch"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, AWARD, MUSICAL_ARTIST\n\nText:\nThe Elephant Man was a huge critical and commercial success, and earned eight Academy Awards nominations, including Academy Award for Best Director and Academy Award for Best Adapted Screenplay for Lynch personally.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, AWARD, MUSICAL_ARTIST\n\nText:\nThe Elephant Man was a huge critical and commercial success, and earned eight Academy Awards nominations, including Academy Award for Best Director and Academy Award for Best Adapted Screenplay for Lynch personally.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_264", "text": "Within months, Hendrix earned three UK top ten hits with the Jimi Hendrix Experience: Hey Joe, Purple Haze, and The Wind Cries Mary.", "requested_types": ["EVENT", "MUSICAL_ARTIST", "COUNTRY", "LOCATION", "BAND", "SONG"], "distractors": ["EVENT", "LOCATION"], "gold": {"EVENT": [], "MUSICAL_ARTIST": ["Hendrix"], "COUNTRY": ["UK"], "LOCATION": [], "BAND": ["the Jimi Hendrix Experience"], "SONG": ["Hey Joe", "Purple Haze", "The Wind Cries Mary"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, COUNTRY, LOCATION, BAND, SONG\n\nText:\nWithin months, Hendrix earned three UK top ten hits with the Jimi Hendrix Experience: Hey Joe, Purple Haze, and The Wind Cries Mary.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, COUNTRY, LOCATION, BAND, SONG\n\nText:\nWithin months, Hendrix earned three UK top ten hits with the Jimi Hendrix Experience: Hey Joe, Purple Haze, and The Wind Cries Mary.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_265", "text": "The set includes the first round of the remastered series plus the long-awaited remastered versions of On Your Feet or on Your Knees (1975), Mirrors, Cultösaurus Erectus, Fire Of Unknown Origin, Extraterrestrial Live, The Revölution by Night, Club Ninja and Imaginos.", "requested_types": ["EVENT", "ALBUM"], "distractors": ["EVENT"], "gold": {"EVENT": [], "ALBUM": ["On Your Feet or on Your Knees", "Mirrors", "Cultösaurus Erectus", "Fire Of Unknown Origin", "Extraterrestrial Live", "The Revölution by Night", "Club Ninja", "Imaginos"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nThe set includes the first round of the remastered series plus the long-awaited remastered versions of On Your Feet or on Your Knees (1975), Mirrors, Cultösaurus Erectus, Fire Of Unknown Origin, Extraterrestrial Live, The Revölution by Night, Club Ninja and Imaginos.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nThe set includes the first round of the remastered series plus the long-awaited remastered versions of On Your Feet or on Your Knees (1975), Mirrors, Cultösaurus Erectus, Fire Of Unknown Origin, Extraterrestrial Live, The Revölution by Night, Club Ninja and Imaginos.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_266", "text": "The musical debuted in San Diego at the Old Globe Theatre in 1986 and premiered on Broadway on November 5, 1987, where it won several Tony Award s, including Tony Award for Best Original Score, Tony Award for Best Book of a Musical, and Best Actress in a Musical (Joanna Gleason), in a year dominated by The Phantom of the Opera (1988).", "requested_types": ["ORGANISATION", "AWARD", "LOCATION", "MUSICAL_INSTRUMENT", "MUSICAL_ARTIST", "MISC"], "distractors": ["MUSICAL_INSTRUMENT"], "gold": {"ORGANISATION": ["Old Globe Theatre", "Broadway"], "AWARD": ["Tony Award", "Tony Award for Best Original Score", "Tony Award for Best Book of a Musical", "Best Actress in a Musical"], "LOCATION": ["San Diego"], "MUSICAL_INSTRUMENT": [], "MUSICAL_ARTIST": ["Joanna Gleason"], "MISC": ["The Phantom of the Opera"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, MISC\n\nText:\nThe musical debuted in San Diego at the Old Globe Theatre in 1986 and premiered on Broadway on November 5, 1987, where it won several Tony Award s, including Tony Award for Best Original Score, Tony Award for Best Book of a Musical, and Best Actress in a Musical (Joanna Gleason), in a year dominated by The Phantom of the Opera (1988).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, MISC\n\nText:\nThe musical debuted in San Diego at the Old Globe Theatre in 1986 and premiered on Broadway on November 5, 1987, where it won several Tony Award s, including Tony Award for Best Original Score, Tony Award for Best Book of a Musical, and Best Actress in a Musical (Joanna Gleason), in a year dominated by The Phantom of the Opera (1988).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_267", "text": "He has won five Primetime Emmy Award s, four Golden Globe Awards, a Grammy Award, and two Screen Actors Guild Awards.", "requested_types": ["COUNTRY", "AWARD", "MUSIC_GENRE"], "distractors": ["MUSIC_GENRE", "COUNTRY"], "gold": {"COUNTRY": [], "AWARD": ["Primetime Emmy Award", "Golden Globe Awards", "Grammy Award", "Screen Actors Guild Awards"], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE\n\nText:\nHe has won five Primetime Emmy Award s, four Golden Globe Awards, a Grammy Award, and two Screen Actors Guild 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: COUNTRY, AWARD, MUSIC_GENRE\n\nText:\nHe has won five Primetime Emmy Award s, four Golden Globe Awards, a Grammy Award, and two Screen Actors Guild 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_music_268", "text": "Simon Fuller has championed many good causes through his working life including The Prince's Trust, Amnesty International, Save the Children, United Nations Foundation, UNICEF, Greenpeace, World Wide Fund for Nature and Comic Relief.", "requested_types": ["BAND", "MUSIC_GENRE", "ORGANISATION", "PERSON"], "distractors": ["BAND", "MUSIC_GENRE"], "gold": {"BAND": [], "MUSIC_GENRE": [], "ORGANISATION": ["The Prince's Trust", "Amnesty International", "Save the Children", "United Nations Foundation", "UNICEF", "Greenpeace", "World Wide Fund for Nature", "Comic Relief"], "PERSON": ["Simon Fuller"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, ORGANISATION, PERSON\n\nText:\nSimon Fuller has championed many good causes through his working life including The Prince's Trust, Amnesty International, Save the Children, United Nations Foundation, UNICEF, Greenpeace, World Wide Fund for Nature and Comic Relief.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, ORGANISATION, PERSON\n\nText:\nSimon Fuller has championed many good causes through his working life including The Prince's Trust, Amnesty International, Save the Children, United Nations Foundation, UNICEF, Greenpeace, World Wide Fund for Nature and Comic Relief.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_269", "text": "Early in their recording careers, Hall and Oates had trouble clearly defining their sound, alternating among Folk music, Soul music, rock and pop.", "requested_types": ["COUNTRY", "BAND", "MUSIC_GENRE"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "BAND": ["Hall and Oates"], "MUSIC_GENRE": ["Folk music", "Soul music", "rock", "pop"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSIC_GENRE\n\nText:\nEarly in their recording careers, Hall and Oates had trouble clearly defining their sound, alternating among Folk music, Soul music, rock and pop.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSIC_GENRE\n\nText:\nEarly in their recording careers, Hall and Oates had trouble clearly defining their sound, alternating among Folk music, Soul music, rock and pop.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_270", "text": "The mid 2000s, especially the United Kingdom and the rest of Europe, saw the continued longevity of nineties boy bands such as Backstreet Boys and Westlife (before they disbanded in 2012), and the successful comeback of Take That in 2005, Boyzone in 2007, and New Kids on the Block in 2008.", "requested_types": ["ALBUM", "BAND", "MUSICAL_ARTIST", "COUNTRY", "LOCATION"], "distractors": ["ALBUM", "MUSICAL_ARTIST"], "gold": {"ALBUM": [], "BAND": ["Backstreet Boys", "Westlife", "Take That", "Boyzone", "New Kids on the Block"], "MUSICAL_ARTIST": [], "COUNTRY": ["United Kingdom"], "LOCATION": ["Europe"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, BAND, MUSICAL_ARTIST, COUNTRY, LOCATION\n\nText:\nThe mid 2000s, especially the United Kingdom and the rest of Europe, saw the continued longevity of nineties boy bands such as Backstreet Boys and Westlife (before they disbanded in 2012), and the successful comeback of Take That in 2005, Boyzone in 2007, and New Kids on the Block in 2008.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, BAND, MUSICAL_ARTIST, COUNTRY, LOCATION\n\nText:\nThe mid 2000s, especially the United Kingdom and the rest of Europe, saw the continued longevity of nineties boy bands such as Backstreet Boys and Westlife (before they disbanded in 2012), and the successful comeback of Take That in 2005, Boyzone in 2007, and New Kids on the Block in 2008.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_271", "text": "The film received Academy Awards nominations for Academy Award for Best Picture (producer Lawrence Turman), Academy Award for Best Actor (Dustin Hoffman), Academy Award for Best Actress (Anne Bancroft), Academy Award for Best Supporting Actress (Katharine Ross), Academy Award for Best Adapted Screenplay (Buck Henry and Calder Willingham), and Academy Award for Best Cinematography (Robert L. Surtees).", "requested_types": ["PERSON", "AWARD", "ALBUM"], "distractors": ["ALBUM"], "gold": {"PERSON": ["Lawrence Turman", "Dustin Hoffman", "Anne Bancroft", "Katharine Ross", "Buck Henry", "Calder Willingham", "Robert L. Surtees"], "AWARD": ["Academy Awards", "Academy Award for Best Picture", "Academy Award for Best Actor", "Academy Award for Best Actress", "Academy Award for Best Supporting Actress", "Academy Award for Best Adapted Screenplay", "Academy Award for Best Cinematography"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nThe film received Academy Awards nominations for Academy Award for Best Picture (producer Lawrence Turman), Academy Award for Best Actor (Dustin Hoffman), Academy Award for Best Actress (Anne Bancroft), Academy Award for Best Supporting Actress (Katharine Ross), Academy Award for Best Adapted Screenplay (Buck Henry and Calder Willingham), and Academy Award for Best Cinematography (Robert L. Surtees).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nThe film received Academy Awards nominations for Academy Award for Best Picture (producer Lawrence Turman), Academy Award for Best Actor (Dustin Hoffman), Academy Award for Best Actress (Anne Bancroft), Academy Award for Best Supporting Actress (Katharine Ross), Academy Award for Best Adapted Screenplay (Buck Henry and Calder Willingham), and Academy Award for Best Cinematography (Robert L. Surtees).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_272", "text": "The album was produced Mike Chapman, also known for his work with The Sweet, Mud, Suzi Quatro, Blondie, Pat Benatar and The Knack.", "requested_types": ["BAND", "MUSIC_GENRE", "AWARD", "MISC", "MUSICAL_ARTIST"], "distractors": ["MUSIC_GENRE", "AWARD", "MISC"], "gold": {"BAND": ["The Sweet", "Mud", "Blondie", "The Knack"], "MUSIC_GENRE": [], "AWARD": [], "MISC": [], "MUSICAL_ARTIST": ["Mike Chapman", "Suzi Quatro", "Pat Benatar"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, AWARD, MISC, MUSICAL_ARTIST\n\nText:\nThe album was produced Mike Chapman, also known for his work with The Sweet, Mud, Suzi Quatro, Blondie, Pat Benatar and The Knack.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, AWARD, MISC, MUSICAL_ARTIST\n\nText:\nThe album was produced Mike Chapman, also known for his work with The Sweet, Mud, Suzi Quatro, Blondie, Pat Benatar and The Knack.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_273", "text": "A few of those include bagad (Breton pipe bands), Fairport Convention, Pentangle, Steeleye Span and Horslips.", "requested_types": ["BAND", "EVENT", "MISC", "MUSICAL_INSTRUMENT"], "distractors": ["MUSICAL_INSTRUMENT", "EVENT"], "gold": {"BAND": ["bagad", "Fairport Convention", "Pentangle", "Steeleye Span", "Horslips"], "EVENT": [], "MISC": ["Breton pipe bands"], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, EVENT, MISC, MUSICAL_INSTRUMENT\n\nText:\nA few of those include bagad (Breton pipe bands), Fairport Convention, Pentangle, Steeleye Span and Horslips.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, EVENT, MISC, MUSICAL_INSTRUMENT\n\nText:\nA few of those include bagad (Breton pipe bands), Fairport Convention, Pentangle, Steeleye Span and Horslips.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_274", "text": "During this period, he also held a variety of positions at the Boston Symphony Orchestra 's summer home in Tanglewood, serving as director of new music activities from 1965 to 1969 and as artistic director of the Tanglewood Music Center from 1970 to 1984 and creating the Tanglewood Festival of Contemporary Music.", "requested_types": ["ORGANISATION", "LOCATION", "ALBUM", "EVENT", "PERSON", "BAND", "SONG"], "distractors": ["SONG", "ALBUM", "PERSON"], "gold": {"ORGANISATION": ["Tanglewood Music Center"], "LOCATION": ["Tanglewood"], "ALBUM": [], "EVENT": ["Tanglewood Festival of Contemporary Music"], "PERSON": [], "BAND": ["Boston Symphony Orchestra"], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, EVENT, PERSON, BAND, SONG\n\nText:\nDuring this period, he also held a variety of positions at the Boston Symphony Orchestra 's summer home in Tanglewood, serving as director of new music activities from 1965 to 1969 and as artistic director of the Tanglewood Music Center from 1970 to 1984 and creating the Tanglewood Festival of Contemporary Music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, EVENT, PERSON, BAND, SONG\n\nText:\nDuring this period, he also held a variety of positions at the Boston Symphony Orchestra 's summer home in Tanglewood, serving as director of new music activities from 1965 to 1969 and as artistic director of the Tanglewood Music Center from 1970 to 1984 and creating the Tanglewood Festival of Contemporary Music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_275", "text": "The British and blues musicians of the early 1960s inspired a number of American blues rock fusion performers, including the The Doors, Canned Heat, the early Jefferson Airplane, Janis Joplin, Johnny Winter, The J. Geils Band, Ry Cooder, and the Allman Brothers Band.", "requested_types": ["PERSON", "BAND", "LOCATION", "SONG", "MUSICAL_ARTIST", "MUSIC_GENRE", "MISC"], "distractors": ["SONG", "LOCATION"], "gold": {"PERSON": ["Janis Joplin"], "BAND": ["The Doors", "Canned Heat", "Jefferson Airplane", "The J. Geils Band", "Allman Brothers Band"], "LOCATION": [], "SONG": [], "MUSICAL_ARTIST": ["Johnny Winter", "Ry Cooder"], "MUSIC_GENRE": ["blues", "blues rock fusion"], "MISC": ["British", "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: PERSON, BAND, LOCATION, SONG, MUSICAL_ARTIST, MUSIC_GENRE, MISC\n\nText:\nThe British and blues musicians of the early 1960s inspired a number of American blues rock fusion performers, including the The Doors, Canned Heat, the early Jefferson Airplane, Janis Joplin, Johnny Winter, The J. Geils Band, Ry Cooder, and the Allman Brothers Band.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, LOCATION, SONG, MUSICAL_ARTIST, MUSIC_GENRE, MISC\n\nText:\nThe British and blues musicians of the early 1960s inspired a number of American blues rock fusion performers, including the The Doors, Canned Heat, the early Jefferson Airplane, Janis Joplin, Johnny Winter, The J. Geils Band, Ry Cooder, and the Allman Brothers Band.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_276", "text": "Other acts who became prominent in the alt-country genre during the 1990s and 2000s included The Bottle Rockets, The Handsome Family, Blue Mountain, Robbie Fulks, Blood Oranges, Bright Eyes, Drive-By Truckers, Old 97's, Old Crow Medicine Show, Nickel Creek, Neko Case, and Whiskeytown, whose lead singer Ryan Adams later had a successful solo-career.", "requested_types": ["BAND", "AWARD", "MUSIC_GENRE", "MUSICAL_ARTIST", "ALBUM"], "distractors": ["AWARD", "ALBUM"], "gold": {"BAND": ["The Bottle Rockets", "The Handsome Family", "Blue Mountain", "Robbie Fulks", "Blood Oranges", "Bright Eyes", "Drive-By Truckers", "Old 97", "Old Crow Medicine Show", "Nickel Creek", "Whiskeytown"], "AWARD": [], "MUSIC_GENRE": ["alt-country"], "MUSICAL_ARTIST": ["Neko Case", "Ryan Adams"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, AWARD, MUSIC_GENRE, MUSICAL_ARTIST, ALBUM\n\nText:\nOther acts who became prominent in the alt-country genre during the 1990s and 2000s included The Bottle Rockets, The Handsome Family, Blue Mountain, Robbie Fulks, Blood Oranges, Bright Eyes, Drive-By Truckers, Old 97's, Old Crow Medicine Show, Nickel Creek, Neko Case, and Whiskeytown, whose lead singer Ryan Adams later had a successful solo-career.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, AWARD, MUSIC_GENRE, MUSICAL_ARTIST, ALBUM\n\nText:\nOther acts who became prominent in the alt-country genre during the 1990s and 2000s included The Bottle Rockets, The Handsome Family, Blue Mountain, Robbie Fulks, Blood Oranges, Bright Eyes, Drive-By Truckers, Old 97's, Old Crow Medicine Show, Nickel Creek, Neko Case, and Whiskeytown, whose lead singer Ryan Adams later had a successful solo-career.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_277", "text": "Respected for her versatility, she received an Academy Juvenile Award, a Golden Globe Award, a Special Tony Award, and was the first woman to win the Grammy Award for Album of the Year for her 1961 live recording Judy at Carnegie Hall.", "requested_types": ["AWARD", "PERSON", "ALBUM"], "distractors": ["PERSON"], "gold": {"AWARD": ["Academy Juvenile Award", "Golden Globe Award", "Special Tony Award", "Grammy Award for Album of the Year"], "PERSON": [], "ALBUM": ["Judy at Carnegie Hall"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nRespected for her versatility, she received an Academy Juvenile Award, a Golden Globe Award, a Special Tony Award, and was the first woman to win the Grammy Award for Album of the Year for her 1961 live recording Judy at Carnegie Hall.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nRespected for her versatility, she received an Academy Juvenile Award, a Golden Globe Award, a Special Tony Award, and was the first woman to win the Grammy Award for Album of the Year for her 1961 live recording Judy at Carnegie Hall.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_278", "text": "However, the 2001 release Destroyer of Worlds was a transitional release that led to a full return to the Viking metal style with the releases of Nordland I (2002) and Nordland II (2003).", "requested_types": ["ALBUM", "MUSIC_GENRE", "MUSICAL_INSTRUMENT"], "distractors": ["MUSICAL_INSTRUMENT"], "gold": {"ALBUM": ["Destroyer of Worlds", "Nordland I", "Nordland II"], "MUSIC_GENRE": ["Viking metal"], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSIC_GENRE, MUSICAL_INSTRUMENT\n\nText:\nHowever, the 2001 release Destroyer of Worlds was a transitional release that led to a full return to the Viking metal style with the releases of Nordland I (2002) and Nordland II (2003).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSIC_GENRE, MUSICAL_INSTRUMENT\n\nText:\nHowever, the 2001 release Destroyer of Worlds was a transitional release that led to a full return to the Viking metal style with the releases of Nordland I (2002) and Nordland II (2003).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_279", "text": "Black Messiah, Beyoncé 's self-titled album (2013), Run the Jewels' Run the Jewels 2 (2014), and Kendrick Lamar 's To Pimp a Butterfly (2015) were noted as laying the groundwork down for the politically charged releases that happened in 2016, which included Rihanna' s Anti, Kanye West's The Life of Pablo, and Beyonce's Formation.", "requested_types": ["LOCATION", "SONG", "MUSICAL_ARTIST", "BAND", "ALBUM"], "distractors": ["LOCATION", "SONG"], "gold": {"LOCATION": [], "SONG": [], "MUSICAL_ARTIST": ["Beyoncé", "Kendrick Lamar", "Rihanna", "Kanye West", "Beyonce"], "BAND": ["Run the Jewels"], "ALBUM": ["Black Messiah", "Run the Jewels 2", "To Pimp a Butterfly", "Anti", "The Life of Pablo", "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: LOCATION, SONG, MUSICAL_ARTIST, BAND, ALBUM\n\nText:\nBlack Messiah, Beyoncé 's self-titled album (2013), Run the Jewels' Run the Jewels 2 (2014), and Kendrick Lamar 's To Pimp a Butterfly (2015) were noted as laying the groundwork down for the politically charged releases that happened in 2016, which included Rihanna' s Anti, Kanye West's The Life of Pablo, and Beyonce's Formation.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST, BAND, ALBUM\n\nText:\nBlack Messiah, Beyoncé 's self-titled album (2013), Run the Jewels' Run the Jewels 2 (2014), and Kendrick Lamar 's To Pimp a Butterfly (2015) were noted as laying the groundwork down for the politically charged releases that happened in 2016, which included Rihanna' s Anti, Kanye West's The Life of Pablo, and Beyonce's Formation.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_280", "text": "With the release of chart-topping albums such as Heart Like a Wheel, Simple Dreams, and Living in the USA, Ronstadt became the first female arena class rock star.", "requested_types": ["MUSICAL_INSTRUMENT", "MUSICAL_ARTIST", "MISC", "AWARD", "ALBUM"], "distractors": ["AWARD", "MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_INSTRUMENT": [], "MUSICAL_ARTIST": ["Ronstadt"], "MISC": ["rock star"], "AWARD": [], "ALBUM": ["Heart Like a Wheel", "Simple Dreams", "Living in the USA"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSICAL_ARTIST, MISC, AWARD, ALBUM\n\nText:\nWith the release of chart-topping albums such as Heart Like a Wheel, Simple Dreams, and Living in the USA, Ronstadt became the first female arena class rock 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: MUSICAL_INSTRUMENT, MUSICAL_ARTIST, MISC, AWARD, ALBUM\n\nText:\nWith the release of chart-topping albums such as Heart Like a Wheel, Simple Dreams, and Living in the USA, Ronstadt became the first female arena class rock 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_music_281", "text": "With Nevermind The success of Nevermind provided numerous Seattle bands, such as Alice in Chains, Pearl Jam, and Soundgarden, access to wider audiences.", "requested_types": ["AWARD", "BAND", "LOCATION", "ALBUM"], "distractors": ["AWARD"], "gold": {"AWARD": [], "BAND": ["Alice in Chains", "Pearl Jam", "Soundgarden"], "LOCATION": ["Seattle"], "ALBUM": ["Nevermind"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, LOCATION, ALBUM\n\nText:\nWith Nevermind The success of Nevermind provided numerous Seattle bands, such as Alice in Chains, Pearl Jam, and Soundgarden, access to wider audiences.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, LOCATION, ALBUM\n\nText:\nWith Nevermind The success of Nevermind provided numerous Seattle bands, such as Alice in Chains, Pearl Jam, and Soundgarden, access to wider audiences.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_282", "text": "Since 2001, the dominance of traditional boy bands on pop charts began to fade in the western hemisphere, although Gil Kaufman of MTV has described new boy bands that are more likely to resemble My Chemical Romance, Sum 41, and Simple Plan.", "requested_types": ["BAND", "ORGANISATION", "ALBUM", "MUSIC_GENRE", "MISC", "PERSON"], "distractors": ["ALBUM", "MISC"], "gold": {"BAND": ["My Chemical Romance", "Sum 41", "Simple Plan"], "ORGANISATION": ["MTV"], "ALBUM": [], "MUSIC_GENRE": ["pop"], "MISC": [], "PERSON": ["Gil Kaufman"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, ORGANISATION, ALBUM, MUSIC_GENRE, MISC, PERSON\n\nText:\nSince 2001, the dominance of traditional boy bands on pop charts began to fade in the western hemisphere, although Gil Kaufman of MTV has described new boy bands that are more likely to resemble My Chemical Romance, Sum 41, and Simple Plan.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, ORGANISATION, ALBUM, MUSIC_GENRE, MISC, PERSON\n\nText:\nSince 2001, the dominance of traditional boy bands on pop charts began to fade in the western hemisphere, although Gil Kaufman of MTV has described new boy bands that are more likely to resemble My Chemical Romance, Sum 41, and Simple Plan.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_283", "text": "In Finland, there emerged a scene that mixed the first wave black metal style with elements of death metal and grindcore; this included Beherit, Archgoat and Impaled Nazarene, whose debut album Tol Cormpt Norz Norz Norz Rock Hard journalist Wolf-Rüdiger Mühlmann considers a part of war metal's roots.", "requested_types": ["ALBUM", "PERSON", "COUNTRY", "ORGANISATION", "MUSIC_GENRE", "BAND", "MISC"], "distractors": ["MISC"], "gold": {"ALBUM": ["Tol Cormpt Norz Norz Norz"], "PERSON": ["Wolf-Rüdiger Mühlmann"], "COUNTRY": ["Finland"], "ORGANISATION": ["Rock Hard"], "MUSIC_GENRE": ["black metal", "death metal", "grindcore", "war metal"], "BAND": ["Beherit", "Archgoat", "Impaled Nazarene"], "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: ALBUM, PERSON, COUNTRY, ORGANISATION, MUSIC_GENRE, BAND, MISC\n\nText:\nIn Finland, there emerged a scene that mixed the first wave black metal style with elements of death metal and grindcore; this included Beherit, Archgoat and Impaled Nazarene, whose debut album Tol Cormpt Norz Norz Norz Rock Hard journalist Wolf-Rüdiger Mühlmann considers a part of war metal's roots.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, PERSON, COUNTRY, ORGANISATION, MUSIC_GENRE, BAND, MISC\n\nText:\nIn Finland, there emerged a scene that mixed the first wave black metal style with elements of death metal and grindcore; this included Beherit, Archgoat and Impaled Nazarene, whose debut album Tol Cormpt Norz Norz Norz Rock Hard journalist Wolf-Rüdiger Mühlmann considers a part of war metal's roots.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_284", "text": "Taiko have a mythological origin in Japanese folklore, but historical records suggest that taiko were introduced to Japan through Korea n and China cultural influence as early as the 6th century CE.", "requested_types": ["COUNTRY", "MISC", "EVENT", "MUSICAL_INSTRUMENT"], "distractors": ["EVENT"], "gold": {"COUNTRY": ["Japan", "Korea", "China"], "MISC": ["Japanese folklore"], "EVENT": [], "MUSICAL_INSTRUMENT": ["Taiko", "taiko"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT, MUSICAL_INSTRUMENT\n\nText:\nTaiko have a mythological origin in Japanese folklore, but historical records suggest that taiko were introduced to Japan through Korea n and China cultural influence as early as the 6th century CE.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT, MUSICAL_INSTRUMENT\n\nText:\nTaiko have a mythological origin in Japanese folklore, but historical records suggest that taiko were introduced to Japan through Korea n and China cultural influence as early as the 6th century CE.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_285", "text": "Focusing on live performance in Scandinavia, Rednex appeared at the 2005 World Championships in Athletics IAAF World Championships in Athletics, in Helsinki.", "requested_types": ["EVENT", "LOCATION", "ALBUM", "BAND"], "distractors": ["ALBUM"], "gold": {"EVENT": ["2005 World Championships in Athletics", "IAAF World Championships in Athletics"], "LOCATION": ["Scandinavia", "Helsinki"], "ALBUM": [], "BAND": ["Rednex"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, BAND\n\nText:\nFocusing on live performance in Scandinavia, Rednex appeared at the 2005 World Championships in Athletics IAAF World Championships in Athletics, in Helsinki.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, BAND\n\nText:\nFocusing on live performance in Scandinavia, Rednex appeared at the 2005 World Championships in Athletics IAAF World Championships in Athletics, in Helsinki.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_286", "text": "Hunt won the Academy Award for Academy Award for Best Actress for starring as Carol Connelly in the romantic comedy As Good as It Gets (1997), while her portrayal of Cheryl Cohen-Greene in The Sessions (2012), gained her an additional Academy Awards nomination for Academy Award for Best Supporting Actress.", "requested_types": ["AWARD", "PERSON", "LOCATION", "MISC"], "distractors": ["LOCATION"], "gold": {"AWARD": ["Academy Award", "Academy Award for Best Actress", "Academy Awards", "Academy Award for Best Supporting Actress"], "PERSON": ["Hunt", "Carol Connelly", "Cheryl Cohen-Greene"], "LOCATION": [], "MISC": ["As Good as It Gets", "The Sessions"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, MISC\n\nText:\nHunt won the Academy Award for Academy Award for Best Actress for starring as Carol Connelly in the romantic comedy As Good as It Gets (1997), while her portrayal of Cheryl Cohen-Greene in The Sessions (2012), gained her an additional Academy Awards nomination for Academy Award for Best Supporting Actress.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, MISC\n\nText:\nHunt won the Academy Award for Academy Award for Best Actress for starring as Carol Connelly in the romantic comedy As Good as It Gets (1997), while her portrayal of Cheryl Cohen-Greene in The Sessions (2012), gained her an additional Academy Awards nomination for Academy Award for Best Supporting Actress.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_287", "text": "In 1990, she contributed a cover of You Do Something to Me to the Cole Porter tribute / AIDS fundraising album Red Hot + Blue produced by the Red Hot Organization.", "requested_types": ["MISC", "MUSICAL_ARTIST", "ORGANISATION", "LOCATION", "ALBUM"], "distractors": ["LOCATION"], "gold": {"MISC": ["AIDS"], "MUSICAL_ARTIST": ["Cole Porter"], "ORGANISATION": ["Red Hot Organization"], "LOCATION": [], "ALBUM": ["You Do Something to Me", "Red Hot + Blue"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ORGANISATION, LOCATION, ALBUM\n\nText:\nIn 1990, she contributed a cover of You Do Something to Me to the Cole Porter tribute / AIDS fundraising album Red Hot + Blue produced by the Red Hot Organization.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ORGANISATION, LOCATION, ALBUM\n\nText:\nIn 1990, she contributed a cover of You Do Something to Me to the Cole Porter tribute / AIDS fundraising album Red Hot + Blue produced by the Red Hot Organization.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_288", "text": "In Southeast Asia, local boy bands also emerged as a result of the continued success of Korean and Japanese boy bands such as SMAP, Shinhwa, TVXQ!, Arashi, Exile, Super Junior, Big Bang, SHINee, EXO, and BTS.", "requested_types": ["MISC", "COUNTRY", "BAND", "MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "LOCATION"], "distractors": ["COUNTRY", "MUSICAL_ARTIST", "MUSICAL_INSTRUMENT"], "gold": {"MISC": ["Korean", "Japanese"], "COUNTRY": [], "BAND": ["SMAP", "Shinhwa", "TVXQ!", "Arashi", "Exile", "Super Junior", "Big Bang", "SHINee", "EXO", "BTS"], "MUSICAL_ARTIST": [], "MUSICAL_INSTRUMENT": [], "LOCATION": ["Southeast Asia"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, LOCATION\n\nText:\nIn Southeast Asia, local boy bands also emerged as a result of the continued success of Korean and Japanese boy bands such as SMAP, Shinhwa, TVXQ!, Arashi, Exile, Super Junior, Big Bang, SHINee, EXO, and BTS.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, LOCATION\n\nText:\nIn Southeast Asia, local boy bands also emerged as a result of the continued success of Korean and Japanese boy bands such as SMAP, Shinhwa, TVXQ!, Arashi, Exile, Super Junior, Big Bang, SHINee, EXO, and BTS.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_289", "text": "It won two Academy Awards, Academy Award for Best Original Score and Academy Award for Best Original Song for A Whole New World and receiving nominations for Best Song (Friend Like Me), Academy Award for Best Sound Editing (Mark A. Mangini), and Academy Award for Best Sound Mixing (Terry Porter, Mel Metcalfe, David J. Hudson and Doc Kane).", "requested_types": ["MISC", "SONG", "MUSIC_GENRE", "ALBUM", "MUSICAL_ARTIST", "AWARD"], "distractors": ["MISC", "ALBUM", "MUSIC_GENRE"], "gold": {"MISC": [], "SONG": ["A Whole New World", "Friend Like Me"], "MUSIC_GENRE": [], "ALBUM": [], "MUSICAL_ARTIST": ["Mark A. Mangini", "Terry Porter", "Mel Metcalfe", "David J. Hudson", "Doc Kane"], "AWARD": ["Academy Awards", "Academy Award for Best Original Score", "Academy Award for Best Original Song", "Best Song", "Academy Award for Best Sound Editing", "Academy Award for Best Sound Mixing"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE, ALBUM, MUSICAL_ARTIST, AWARD\n\nText:\nIt won two Academy Awards, Academy Award for Best Original Score and Academy Award for Best Original Song for A Whole New World and receiving nominations for Best Song (Friend Like Me), Academy Award for Best Sound Editing (Mark A. Mangini), and Academy Award for Best Sound Mixing (Terry Porter, Mel Metcalfe, David J. Hudson and Doc Kane).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE, ALBUM, MUSICAL_ARTIST, AWARD\n\nText:\nIt won two Academy Awards, Academy Award for Best Original Score and Academy Award for Best Original Song for A Whole New World and receiving nominations for Best Song (Friend Like Me), Academy Award for Best Sound Editing (Mark A. Mangini), and Academy Award for Best Sound Mixing (Terry Porter, Mel Metcalfe, David J. Hudson and Doc Kane).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_290", "text": "Located in downtown Los Angeles, The Music Center is home to the Dorothy Chandler Pavilion, Ahmanson Theater, Mark Taper Forum, Roy and Edna Disney / CalArts Theatre, and Walt Disney Concert Hall.", "requested_types": ["MUSICAL_ARTIST", "EVENT", "LOCATION"], "distractors": ["EVENT", "MUSICAL_ARTIST"], "gold": {"MUSICAL_ARTIST": [], "EVENT": [], "LOCATION": ["Los Angeles", "The Music Center", "Dorothy Chandler Pavilion", "Ahmanson Theater", "Mark Taper Forum", "Roy and Edna Disney / CalArts Theatre", "Walt Disney Concert Hall"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, EVENT, LOCATION\n\nText:\nLocated in downtown Los Angeles, The Music Center is home to the Dorothy Chandler Pavilion, Ahmanson Theater, Mark Taper Forum, Roy and Edna Disney / CalArts Theatre, and Walt Disney Concert Hall.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, EVENT, LOCATION\n\nText:\nLocated in downtown Los Angeles, The Music Center is home to the Dorothy Chandler Pavilion, Ahmanson Theater, Mark Taper Forum, Roy and Edna Disney / CalArts Theatre, and Walt Disney Concert Hall.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_291", "text": "Highlights in 2002 included the premieres of several major works, including View From Olympus a double concerto for piano, percussion and orchestra performed by Evelyn Glennie, Philip Smith and the Halle Orchestra conducted by Mark Elder at the Royal Gala finale of the 2002 Commonwealth Games' Pulse 'music festival in Manchester, UK.", "requested_types": ["COUNTRY", "MUSICAL_INSTRUMENT", "ORGANISATION", "AWARD", "EVENT", "BAND", "LOCATION", "MISC", "PERSON", "ALBUM", "MUSICAL_ARTIST"], "distractors": ["ORGANISATION", "AWARD", "MISC"], "gold": {"COUNTRY": ["UK"], "MUSICAL_INSTRUMENT": ["piano"], "ORGANISATION": [], "AWARD": [], "EVENT": ["Royal Gala finale", "2002 Commonwealth Games", "Pulse 'music festival"], "BAND": ["Halle Orchestra"], "LOCATION": ["Manchester"], "MISC": [], "PERSON": ["Mark Elder"], "ALBUM": ["View From Olympus"], "MUSICAL_ARTIST": ["Evelyn Glennie", "Philip Smith"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, ORGANISATION, AWARD, EVENT, BAND, LOCATION, MISC, PERSON, ALBUM, MUSICAL_ARTIST\n\nText:\nHighlights in 2002 included the premieres of several major works, including View From Olympus a double concerto for piano, percussion and orchestra performed by Evelyn Glennie, Philip Smith and the Halle Orchestra conducted by Mark Elder at the Royal Gala finale of the 2002 Commonwealth Games' Pulse 'music festival in Manchester, UK.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, ORGANISATION, AWARD, EVENT, BAND, LOCATION, MISC, PERSON, ALBUM, MUSICAL_ARTIST\n\nText:\nHighlights in 2002 included the premieres of several major works, including View From Olympus a double concerto for piano, percussion and orchestra performed by Evelyn Glennie, Philip Smith and the Halle Orchestra conducted by Mark Elder at the Royal Gala finale of the 2002 Commonwealth Games' Pulse 'music festival in Manchester, UK.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_292", "text": "Rock and roll songs critical of disco included Bob Seger 's Old Time Rock and Roll and, especially, The Who' s Sister Disco (both 1978) - although The Who's Eminence Front (four years later) had a disco feel.", "requested_types": ["SONG", "MUSICAL_ARTIST", "PERSON", "BAND", "MUSIC_GENRE"], "distractors": ["PERSON"], "gold": {"SONG": ["Old Time Rock and Roll", "Sister Disco", "Eminence Front"], "MUSICAL_ARTIST": ["Bob Seger"], "PERSON": [], "BAND": ["The Who"], "MUSIC_GENRE": ["Rock and roll", "disco"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, PERSON, BAND, MUSIC_GENRE\n\nText:\nRock and roll songs critical of disco included Bob Seger 's Old Time Rock and Roll and, especially, The Who' s Sister Disco (both 1978) - although The Who's Eminence Front (four years later) had a disco feel.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, PERSON, BAND, MUSIC_GENRE\n\nText:\nRock and roll songs critical of disco included Bob Seger 's Old Time Rock and Roll and, especially, The Who' s Sister Disco (both 1978) - although The Who's Eminence Front (four years later) had a disco feel.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_293", "text": "These have been used in clay shooting and were suggested for use in the Modern pentathlon at the 2012 Summer Olympics after a successful trial at the 2010 Summer Youth Olympics in Singapore.", "requested_types": ["MUSICAL_ARTIST", "LOCATION", "COUNTRY", "EVENT"], "distractors": ["LOCATION", "MUSICAL_ARTIST"], "gold": {"MUSICAL_ARTIST": [], "LOCATION": [], "COUNTRY": ["Singapore"], "EVENT": ["Modern pentathlon at the 2012 Summer Olympics", "2010 Summer Youth Olympics"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, LOCATION, COUNTRY, EVENT\n\nText:\nThese have been used in clay shooting and were suggested for use in the Modern pentathlon at the 2012 Summer Olympics after a successful trial at the 2010 Summer Youth Olympics in Singapore.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, LOCATION, COUNTRY, EVENT\n\nText:\nThese have been used in clay shooting and were suggested for use in the Modern pentathlon at the 2012 Summer Olympics after a successful trial at the 2010 Summer Youth Olympics in Singapore.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_294", "text": "On September 12, 2010, during the Chilean bicentennial festivities, President Sebastián Piñera announced that the capacity of the stadium will be increased so as to reach 70,000 seats for the 2014 South American Games that will take place in Santiago.", "requested_types": ["EVENT", "LOCATION", "PERSON", "ALBUM"], "distractors": ["ALBUM"], "gold": {"EVENT": ["Chilean bicentennial festivities", "2014 South American Games"], "LOCATION": ["Santiago"], "PERSON": ["Sebastián Piñera"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, ALBUM\n\nText:\nOn September 12, 2010, during the Chilean bicentennial festivities, President Sebastián Piñera announced that the capacity of the stadium will be increased so as to reach 70,000 seats for the 2014 South American Games that will take place in Santiago.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, ALBUM\n\nText:\nOn September 12, 2010, during the Chilean bicentennial festivities, President Sebastián Piñera announced that the capacity of the stadium will be increased so as to reach 70,000 seats for the 2014 South American Games that will take place in Santiago.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_295", "text": "The subsequent albums The Clones of Dr. Funkenstein (1976), Funkentelechy vs. the Placebo Syndrome (1977), and Motor Booty Affair (1978) all reached high on both the R & amp; B and Pop charts, while Funkadelic was also experiencing significant mainstream success.", "requested_types": ["MISC", "ALBUM", "MUSIC_GENRE", "BAND"], "distractors": ["MISC"], "gold": {"MISC": [], "ALBUM": ["The Clones of Dr. Funkenstein", "Funkentelechy vs. the Placebo Syndrome", "Motor Booty Affair"], "MUSIC_GENRE": ["R & amp; B", "Pop"], "BAND": ["Funkadelic"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MUSIC_GENRE, BAND\n\nText:\nThe subsequent albums The Clones of Dr. Funkenstein (1976), Funkentelechy vs. the Placebo Syndrome (1977), and Motor Booty Affair (1978) all reached high on both the R & amp; B and Pop charts, while Funkadelic was also experiencing significant mainstream success.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MUSIC_GENRE, BAND\n\nText:\nThe subsequent albums The Clones of Dr. Funkenstein (1976), Funkentelechy vs. the Placebo Syndrome (1977), and Motor Booty Affair (1978) all reached high on both the R & amp; B and Pop charts, while Funkadelic was also experiencing significant mainstream success.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_296", "text": "Deep Purple and Whitesnake 's David Coverdale, Samson' s Nicky Moore and Lone Star 's John Sloman were all considered and Iommi states in his autobiography that Michael Bolton auditioned.", "requested_types": ["MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "BAND"], "distractors": ["MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_ARTIST": ["David Coverdale", "Nicky Moore", "John Sloman", "Iommi", "Michael Bolton"], "MUSICAL_INSTRUMENT": [], "BAND": ["Deep Purple", "Whitesnake", "Samson", "Lone Star"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MUSICAL_INSTRUMENT, BAND\n\nText:\nDeep Purple and Whitesnake 's David Coverdale, Samson' s Nicky Moore and Lone Star 's John Sloman were all considered and Iommi states in his autobiography that Michael Bolton auditioned.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MUSICAL_INSTRUMENT, BAND\n\nText:\nDeep Purple and Whitesnake 's David Coverdale, Samson' s Nicky Moore and Lone Star 's John Sloman were all considered and Iommi states in his autobiography that Michael Bolton auditioned.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_297", "text": "Indie band Rilo Kiley, in keeping with their tendency to explore a variety of rockish styles, incorporated funk into their song The Moneymaker on the album Under the Blacklight.", "requested_types": ["MISC", "ORGANISATION", "MUSIC_GENRE", "ALBUM", "MUSICAL_ARTIST", "BAND", "AWARD", "SONG"], "distractors": ["ORGANISATION", "AWARD", "MUSICAL_ARTIST"], "gold": {"MISC": ["rockish"], "ORGANISATION": [], "MUSIC_GENRE": ["funk"], "ALBUM": ["Under the Blacklight"], "MUSICAL_ARTIST": [], "BAND": ["Rilo Kiley"], "AWARD": [], "SONG": ["The Moneymaker"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, ALBUM, MUSICAL_ARTIST, BAND, AWARD, SONG\n\nText:\nIndie band Rilo Kiley, in keeping with their tendency to explore a variety of rockish styles, incorporated funk into their song The Moneymaker on the album Under the Blacklight.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, ALBUM, MUSICAL_ARTIST, BAND, AWARD, SONG\n\nText:\nIndie band Rilo Kiley, in keeping with their tendency to explore a variety of rockish styles, incorporated funk into their song The Moneymaker on the album Under the Blacklight.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_298", "text": "The box set was accompanied by 5 Album Studio Set, which contains only the first five studio albums (excluding Cut the Crap), and The Clash Hits Back, a 33-track, two-CD best of collection sequenced to copy the set played by the band at the Brixton Fair Deal (now the Academy) on 19 July 1982.", "requested_types": ["COUNTRY", "LOCATION", "ORGANISATION", "ALBUM", "MUSICAL_ARTIST"], "distractors": ["LOCATION", "COUNTRY", "MUSICAL_ARTIST"], "gold": {"COUNTRY": [], "LOCATION": [], "ORGANISATION": ["Brixton Fair Deal"], "ALBUM": ["5 Album Studio Set", "Cut the Crap", "The Clash Hits Back"], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ORGANISATION, ALBUM, MUSICAL_ARTIST\n\nText:\nThe box set was accompanied by 5 Album Studio Set, which contains only the first five studio albums (excluding Cut the Crap), and The Clash Hits Back, a 33-track, two-CD best of collection sequenced to copy the set played by the band at the Brixton Fair Deal (now the Academy) on 19 July 1982.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ORGANISATION, ALBUM, MUSICAL_ARTIST\n\nText:\nThe box set was accompanied by 5 Album Studio Set, which contains only the first five studio albums (excluding Cut the Crap), and The Clash Hits Back, a 33-track, two-CD best of collection sequenced to copy the set played by the band at the Brixton Fair Deal (now the Academy) on 19 July 1982.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_299", "text": "Six albums released by the label have been certified gold for sales of 500,000 copies: Oh, Inverted World, Chutes Too Narrow and Wincing the Night Away, all by The Shins, Fleet Foxes by Fleet Foxes, The Head and the Heart by The Head and the Heart, and Everything All the Time by Band of Horses.", "requested_types": ["ALBUM", "BAND", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_ARTIST"], "gold": {"ALBUM": ["Oh, Inverted World", "Chutes Too Narrow", "Wincing the Night Away", "Everything All the Time"], "BAND": ["The Shins", "Fleet Foxes", "The Head and the Heart", "Band of Horses"], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, BAND, MUSICAL_ARTIST\n\nText:\nSix albums released by the label have been certified gold for sales of 500,000 copies: Oh, Inverted World, Chutes Too Narrow and Wincing the Night Away, all by The Shins, Fleet Foxes by Fleet Foxes, The Head and the Heart by The Head and the Heart, and Everything All the Time by Band of Horses.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, BAND, MUSICAL_ARTIST\n\nText:\nSix albums released by the label have been certified gold for sales of 500,000 copies: Oh, Inverted World, Chutes Too Narrow and Wincing the Night Away, all by The Shins, Fleet Foxes by Fleet Foxes, The Head and the Heart by The Head and the Heart, and Everything All the Time by Band of Horses.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_300", "text": "In the Southwestern United States, it was the Rocky Mountains, American frontier, and Rio Grande that acted as a similar backdrop for Indigenous music of North America, Mexican, and cowboy ballads, which resulted in New Mexico music and the development of Western music, and its directly related Red Dirt, Texas country music, and Tejano music styles.", "requested_types": ["MISC", "MUSICAL_INSTRUMENT", "LOCATION", "MUSIC_GENRE", "BAND"], "distractors": ["MUSICAL_INSTRUMENT", "BAND"], "gold": {"MISC": ["Mexican", "cowboy ballads", "Western music"], "MUSICAL_INSTRUMENT": [], "LOCATION": ["Southwestern United States", "Rocky Mountains", "American frontier", "Rio Grande"], "MUSIC_GENRE": ["Indigenous music of North America", "New Mexico music", "Red Dirt", "Texas country music", "Tejano music"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, LOCATION, MUSIC_GENRE, BAND\n\nText:\nIn the Southwestern United States, it was the Rocky Mountains, American frontier, and Rio Grande that acted as a similar backdrop for Indigenous music of North America, Mexican, and cowboy ballads, which resulted in New Mexico music and the development of Western music, and its directly related Red Dirt, Texas country music, and Tejano music styles.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, LOCATION, MUSIC_GENRE, BAND\n\nText:\nIn the Southwestern United States, it was the Rocky Mountains, American frontier, and Rio Grande that acted as a similar backdrop for Indigenous music of North America, Mexican, and cowboy ballads, which resulted in New Mexico music and the development of Western music, and its directly related Red Dirt, Texas country music, and Tejano music styles.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_301", "text": "The picture received Academy Awards nominations for Academy Award for Best Sound Mixing (Loren L. Ryder) and Academy Award for Best Original Screenplay.", "requested_types": ["EVENT", "MUSICAL_ARTIST", "AWARD", "MUSICAL_INSTRUMENT"], "distractors": ["EVENT", "MUSICAL_INSTRUMENT"], "gold": {"EVENT": [], "MUSICAL_ARTIST": ["Loren L. Ryder"], "AWARD": ["Academy Awards", "Academy Award for Best Sound Mixing", "Academy Award for Best Original Screenplay"], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, AWARD, MUSICAL_INSTRUMENT\n\nText:\nThe picture received Academy Awards nominations for Academy Award for Best Sound Mixing (Loren L. Ryder) and Academy Award for Best Original Screenplay.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, AWARD, MUSICAL_INSTRUMENT\n\nText:\nThe picture received Academy Awards nominations for Academy Award for Best Sound Mixing (Loren L. Ryder) and Academy Award for Best Original Screenplay.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_302", "text": "By the mid-1980s, bands began proliferating and became increasingly popular, including the Sisters of Mercy, the Mission, Alien Sex Fiend, the March Violets, Xmal Deutschland, the Membranes, and Fields of the Nephilim.", "requested_types": ["BAND", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"BAND": ["the Sisters of Mercy", "the Mission", "Alien Sex Fiend", "the March Violets", "Xmal Deutschland", "the Membranes", "Fields of the Nephilim"], "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: BAND, COUNTRY\n\nText:\nBy the mid-1980s, bands began proliferating and became increasingly popular, including the Sisters of Mercy, the Mission, Alien Sex Fiend, the March Violets, Xmal Deutschland, the Membranes, and Fields of the Nephilim.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, COUNTRY\n\nText:\nBy the mid-1980s, bands began proliferating and became increasingly popular, including the Sisters of Mercy, the Mission, Alien Sex Fiend, the March Violets, Xmal Deutschland, the Membranes, and Fields of the Nephilim.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_303", "text": "Breakbeats have been used in styles such as hip hop, Jungle music, drum and bass, big beat, hardcore, and UK garage styles (including 2-step garage, breakstep and dubstep).", "requested_types": ["EVENT", "LOCATION", "MISC", "MUSIC_GENRE"], "distractors": ["EVENT", "LOCATION"], "gold": {"EVENT": [], "LOCATION": [], "MISC": ["Breakbeats"], "MUSIC_GENRE": ["hip hop", "Jungle music", "drum and bass", "big beat", "hardcore", "UK garage", "2-step garage", "breakstep", "dubstep"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, MUSIC_GENRE\n\nText:\nBreakbeats have been used in styles such as hip hop, Jungle music, drum and bass, big beat, hardcore, and UK garage styles (including 2-step garage, breakstep and dubstep).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, MUSIC_GENRE\n\nText:\nBreakbeats have been used in styles such as hip hop, Jungle music, drum and bass, big beat, hardcore, and UK garage styles (including 2-step garage, breakstep and dubstep).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_304", "text": "In 2016 Chuck D joined the band Prophets of Rage along with B-Real and former members of Rage Against the Machine.", "requested_types": ["SONG", "BAND", "MUSIC_GENRE", "MUSICAL_ARTIST", "MUSICAL_INSTRUMENT"], "distractors": ["MUSICAL_INSTRUMENT", "SONG", "MUSIC_GENRE"], "gold": {"SONG": [], "BAND": ["Prophets of Rage", "Rage Against the Machine"], "MUSIC_GENRE": [], "MUSICAL_ARTIST": ["Chuck D", "B-Real"], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, MUSIC_GENRE, MUSICAL_ARTIST, MUSICAL_INSTRUMENT\n\nText:\nIn 2016 Chuck D joined the band Prophets of Rage along with B-Real and former members of Rage Against the Machine.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, MUSIC_GENRE, MUSICAL_ARTIST, MUSICAL_INSTRUMENT\n\nText:\nIn 2016 Chuck D joined the band Prophets of Rage along with B-Real and former members of Rage Against the Machine.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_305", "text": "The viola is also an important accompaniment instrument in Slovakia n, Hungary and Romania n folk string band music, especially in Transylvania.", "requested_types": ["BAND", "MISC", "ALBUM", "COUNTRY", "LOCATION", "MUSICAL_INSTRUMENT"], "distractors": ["MISC", "BAND", "ALBUM"], "gold": {"BAND": [], "MISC": [], "ALBUM": [], "COUNTRY": ["Slovakia", "Hungary", "Romania"], "LOCATION": ["Transylvania"], "MUSICAL_INSTRUMENT": ["viola"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MISC, ALBUM, COUNTRY, LOCATION, MUSICAL_INSTRUMENT\n\nText:\nThe viola is also an important accompaniment instrument in Slovakia n, Hungary and Romania n folk string band music, especially in Transylvania.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MISC, ALBUM, COUNTRY, LOCATION, MUSICAL_INSTRUMENT\n\nText:\nThe viola is also an important accompaniment instrument in Slovakia n, Hungary and Romania n folk string band music, especially in Transylvania.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_306", "text": "Folsom Field is also used as the finish line for the Bolder Boulder, a popular 10K run.", "requested_types": ["EVENT", "LOCATION", "ALBUM", "MISC"], "distractors": ["ALBUM"], "gold": {"EVENT": ["Bolder Boulder"], "LOCATION": ["Folsom Field"], "ALBUM": [], "MISC": ["10K run"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MISC\n\nText:\nFolsom Field is also used as the finish line for the Bolder Boulder, a popular 10K run.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MISC\n\nText:\nFolsom Field is also used as the finish line for the Bolder Boulder, a popular 10K run.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_307", "text": "However, some alt-country songs have been crossover hits to mainstream country radio in cover versions by established artists on the format; Lucinda Williams' Passionate Kisses was a hit for Mary Chapin Carpenter in 1993, Ryan Adams's When The Stars Go Blue was a hit for Tim McGraw in 2007, and Old Crow Medicine Show's Wagon Wheel was a hit for Darius Rucker in 2013.", "requested_types": ["MISC", "EVENT", "MUSIC_GENRE", "MUSICAL_ARTIST", "SONG", "BAND", "LOCATION"], "distractors": ["LOCATION", "EVENT"], "gold": {"MISC": ["mainstream country radio"], "EVENT": [], "MUSIC_GENRE": ["alt-country"], "MUSICAL_ARTIST": ["Lucinda Williams", "Mary Chapin Carpenter", "Ryan Adams", "Tim McGraw", "Darius Rucker"], "SONG": ["Passionate Kisses", "When The Stars Go Blue"], "BAND": ["Old Crow Medicine Show's", "Wagon Wheel"], "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, EVENT, MUSIC_GENRE, MUSICAL_ARTIST, SONG, BAND, LOCATION\n\nText:\nHowever, some alt-country songs have been crossover hits to mainstream country radio in cover versions by established artists on the format; Lucinda Williams' Passionate Kisses was a hit for Mary Chapin Carpenter in 1993, Ryan Adams's When The Stars Go Blue was a hit for Tim McGraw in 2007, and Old Crow Medicine Show's Wagon Wheel was a hit for Darius Rucker in 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: MISC, EVENT, MUSIC_GENRE, MUSICAL_ARTIST, SONG, BAND, LOCATION\n\nText:\nHowever, some alt-country songs have been crossover hits to mainstream country radio in cover versions by established artists on the format; Lucinda Williams' Passionate Kisses was a hit for Mary Chapin Carpenter in 1993, Ryan Adams's When The Stars Go Blue was a hit for Tim McGraw in 2007, and Old Crow Medicine Show's Wagon Wheel was a hit for Darius Rucker in 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_music_308", "text": "During their three years as a mainstream act, Nirvana was awarded an American Music Awards, Brit Awards, Grammy Award, seven MTV Video Music Award s and two NME Awards.", "requested_types": ["BAND", "AWARD", "SONG", "MUSICAL_ARTIST", "EVENT"], "distractors": ["SONG", "MUSICAL_ARTIST", "EVENT"], "gold": {"BAND": ["Nirvana"], "AWARD": ["American Music Awards", "Brit Awards", "Grammy Award", "MTV Video Music Award", "NME Awards"], "SONG": [], "MUSICAL_ARTIST": [], "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: BAND, AWARD, SONG, MUSICAL_ARTIST, EVENT\n\nText:\nDuring their three years as a mainstream act, Nirvana was awarded an American Music Awards, Brit Awards, Grammy Award, seven MTV Video Music Award s and two NME 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: BAND, AWARD, SONG, MUSICAL_ARTIST, EVENT\n\nText:\nDuring their three years as a mainstream act, Nirvana was awarded an American Music Awards, Brit Awards, Grammy Award, seven MTV Video Music Award s and two NME 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_music_309", "text": "This was reflected in a series of albums released by Island Records, including Swordfishtrombones (1983), Rain Dogs (1985), and Franks Wild Years (1987).", "requested_types": ["ORGANISATION", "EVENT", "COUNTRY", "ALBUM"], "distractors": ["COUNTRY", "EVENT"], "gold": {"ORGANISATION": ["Island Records"], "EVENT": [], "COUNTRY": [], "ALBUM": ["Swordfishtrombones", "Rain Dogs", "Franks Wild Years"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, ALBUM\n\nText:\nThis was reflected in a series of albums released by Island Records, including Swordfishtrombones (1983), Rain Dogs (1985), and Franks Wild Years (1987).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, ALBUM\n\nText:\nThis was reflected in a series of albums released by Island Records, including Swordfishtrombones (1983), Rain Dogs (1985), and Franks Wild Years (1987).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_310", "text": "Pink is also involved with several charities, including Human Rights Campaign, ONE Campaign, Prince's Trust, New York Restoration Project, Run for the Cure Foundation, Save the Children, Take Back the Night, UNICEF and World Animal Protection. Look to the Stars: that swept through the Australian state of Victoria earlier that month.", "requested_types": ["PERSON", "LOCATION", "MISC", "MUSICAL_INSTRUMENT", "MUSICAL_ARTIST", "ORGANISATION", "SONG"], "distractors": ["PERSON", "MUSICAL_INSTRUMENT", "SONG"], "gold": {"PERSON": [], "LOCATION": ["Victoria"], "MISC": ["Look to the Stars", "Australian"], "MUSICAL_INSTRUMENT": [], "MUSICAL_ARTIST": ["Pink"], "ORGANISATION": ["Human Rights Campaign", "ONE Campaign", "Prince's Trust", "New York Restoration Project", "Run for the Cure Foundation", "Save the Children", "Take Back the Night", "UNICEF", "World Animal Protection"], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, ORGANISATION, SONG\n\nText:\nPink is also involved with several charities, including Human Rights Campaign, ONE Campaign, Prince's Trust, New York Restoration Project, Run for the Cure Foundation, Save the Children, Take Back the Night, UNICEF and World Animal Protection. Look to the Stars: that swept through the Australian state of Victoria earlier that month.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, MUSICAL_INSTRUMENT, MUSICAL_ARTIST, ORGANISATION, SONG\n\nText:\nPink is also involved with several charities, including Human Rights Campaign, ONE Campaign, Prince's Trust, New York Restoration Project, Run for the Cure Foundation, Save the Children, Take Back the Night, UNICEF and World Animal Protection. Look to the Stars: that swept through the Australian state of Victoria earlier that month.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_311", "text": "The band's first two releases on the label, Enjoy Incubus and S.C.I.E.N.C.E. went largely unnoticed in the mainstream, but subsequent releases Make Yourself and Morning View were commercial successes.", "requested_types": ["ALBUM", "COUNTRY", "AWARD"], "distractors": ["COUNTRY", "AWARD"], "gold": {"ALBUM": ["Enjoy Incubus", "S.C.I.E.N.C.E.", "Make Yourself", "Morning View"], "COUNTRY": [], "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: ALBUM, COUNTRY, AWARD\n\nText:\nThe band's first two releases on the label, Enjoy Incubus and S.C.I.E.N.C.E. went largely unnoticed in the mainstream, but subsequent releases Make Yourself and Morning View were commercial successes.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, COUNTRY, AWARD\n\nText:\nThe band's first two releases on the label, Enjoy Incubus and S.C.I.E.N.C.E. went largely unnoticed in the mainstream, but subsequent releases Make Yourself and Morning View were commercial successes.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_312", "text": "Reggae en Español spread from mainland South American Caribbean from Venezuela and Guyana to the rest of South America.", "requested_types": ["COUNTRY", "LOCATION", "ORGANISATION", "MUSIC_GENRE", "MISC"], "distractors": ["ORGANISATION", "MISC"], "gold": {"COUNTRY": ["Venezuela", "Guyana"], "LOCATION": ["South American", "Caribbean", "South America"], "ORGANISATION": [], "MUSIC_GENRE": ["Reggae en Español"], "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: COUNTRY, LOCATION, ORGANISATION, MUSIC_GENRE, MISC\n\nText:\nReggae en Español spread from mainland South American Caribbean from Venezuela and Guyana to the rest of South 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: COUNTRY, LOCATION, ORGANISATION, MUSIC_GENRE, MISC\n\nText:\nReggae en Español spread from mainland South American Caribbean from Venezuela and Guyana to the rest of South 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_music_313", "text": "Tyler came to prominence with the release of her 1977 album The World Starts Tonight and its singles Lost in France and More Than a Lover.", "requested_types": ["EVENT", "MUSICAL_ARTIST", "PERSON", "ALBUM", "SONG", "MUSICAL_INSTRUMENT"], "distractors": ["EVENT", "PERSON", "MUSICAL_INSTRUMENT"], "gold": {"EVENT": [], "MUSICAL_ARTIST": ["Tyler"], "PERSON": [], "ALBUM": ["The World Starts Tonight"], "SONG": ["Lost in France", "More Than a Lover"], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, PERSON, ALBUM, SONG, MUSICAL_INSTRUMENT\n\nText:\nTyler came to prominence with the release of her 1977 album The World Starts Tonight and its singles Lost in France and More Than a Lover.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, PERSON, ALBUM, SONG, MUSICAL_INSTRUMENT\n\nText:\nTyler came to prominence with the release of her 1977 album The World Starts Tonight and its singles Lost in France and More Than a Lover.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_314", "text": "The group's evolution can be traced through the recordings The John Coltrane Quartet Plays, Living Space and Transition (both June 1965), New Thing at Newport (July 1965), Sun Ship (August 1965), and First Meditations (September 1965).", "requested_types": ["ALBUM", "COUNTRY", "MUSICAL_ARTIST"], "distractors": ["COUNTRY", "MUSICAL_ARTIST"], "gold": {"ALBUM": ["The John Coltrane Quartet Plays", "Living Space", "Transition", "New Thing at Newport", "Sun Ship", "First Meditations"], "COUNTRY": [], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, COUNTRY, MUSICAL_ARTIST\n\nText:\nThe group's evolution can be traced through the recordings The John Coltrane Quartet Plays, Living Space and Transition (both June 1965), New Thing at Newport (July 1965), Sun Ship (August 1965), and First Meditations (September 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: ALBUM, COUNTRY, MUSICAL_ARTIST\n\nText:\nThe group's evolution can be traced through the recordings The John Coltrane Quartet Plays, Living Space and Transition (both June 1965), New Thing at Newport (July 1965), Sun Ship (August 1965), and First Meditations (September 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_music_315", "text": "Female artists such as Reba McEntire, Patty Loveless, Faith Hill, Martina McBride, Deana Carter, LeAnn Rimes, Mindy McCready, Lorrie Morgan, Shania Twain, and Mary Chapin Carpenter all released platinum-selling albums in the 1990s.", "requested_types": ["AWARD", "MUSICAL_ARTIST", "PERSON", "MISC", "MUSICAL_INSTRUMENT"], "distractors": ["MUSICAL_INSTRUMENT", "MISC", "AWARD"], "gold": {"AWARD": [], "MUSICAL_ARTIST": ["Patty Loveless", "Martina McBride", "Deana Carter", "LeAnn Rimes", "Mindy McCready", "Lorrie Morgan", "Mary Chapin Carpenter"], "PERSON": ["Reba McEntire", "Faith Hill", "Shania Twain"], "MISC": [], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, PERSON, MISC, MUSICAL_INSTRUMENT\n\nText:\nFemale artists such as Reba McEntire, Patty Loveless, Faith Hill, Martina McBride, Deana Carter, LeAnn Rimes, Mindy McCready, Lorrie Morgan, Shania Twain, and Mary Chapin Carpenter all released platinum-selling albums in the 1990s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, PERSON, MISC, MUSICAL_INSTRUMENT\n\nText:\nFemale artists such as Reba McEntire, Patty Loveless, Faith Hill, Martina McBride, Deana Carter, LeAnn Rimes, Mindy McCready, Lorrie Morgan, Shania Twain, and Mary Chapin Carpenter all released platinum-selling albums in the 1990s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_316", "text": "On 19 September, he played with Korean artist PSY for opening ceremony of 2014 Asian Games in Incheon, South Korea.", "requested_types": ["LOCATION", "PERSON", "MISC", "MUSICAL_ARTIST", "EVENT", "COUNTRY"], "distractors": ["PERSON"], "gold": {"LOCATION": ["Incheon"], "PERSON": [], "MISC": ["Korean"], "MUSICAL_ARTIST": ["PSY"], "EVENT": ["2014 Asian Games"], "COUNTRY": ["South Korea"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, MUSICAL_ARTIST, EVENT, COUNTRY\n\nText:\nOn 19 September, he played with Korean artist PSY for opening ceremony of 2014 Asian Games in Incheon, South Korea.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, MUSICAL_ARTIST, EVENT, COUNTRY\n\nText:\nOn 19 September, he played with Korean artist PSY for opening ceremony of 2014 Asian Games in Incheon, South Korea.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_317", "text": "The Chess - concept album - with vocals by Elaine Paige, Barbara Dickson, Murray Head and Swedes Tommy Körberg and Björn Skifs - was released in October 1984, selling two million copies worldwide.", "requested_types": ["MISC", "BAND", "ALBUM", "MUSICAL_ARTIST"], "distractors": ["ALBUM", "BAND"], "gold": {"MISC": ["Chess", "concept album", "Swedes"], "BAND": [], "ALBUM": [], "MUSICAL_ARTIST": ["Elaine Paige", "Barbara Dickson", "Murray Head", "Tommy Körberg", "Björn Skifs"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM, MUSICAL_ARTIST\n\nText:\nThe Chess - concept album - with vocals by Elaine Paige, Barbara Dickson, Murray Head and Swedes Tommy Körberg and Björn Skifs - was released in October 1984, selling two million copies 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: MISC, BAND, ALBUM, MUSICAL_ARTIST\n\nText:\nThe Chess - concept album - with vocals by Elaine Paige, Barbara Dickson, Murray Head and Swedes Tommy Körberg and Björn Skifs - was released in October 1984, selling two million copies 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_music_318", "text": "Shore has also been honored with awards from National Board of Review, Recording Academy Honors, Broadcast Film Critics Association, Chicago Film Critics, Genie Award, World Soundtrack Award, New York's Gotham Award, and The Saturn Award for Science Fiction.", "requested_types": ["AWARD", "LOCATION", "MUSICAL_ARTIST", "COUNTRY", "ORGANISATION", "PERSON"], "distractors": ["COUNTRY", "PERSON"], "gold": {"AWARD": ["Recording Academy Honors", "Genie Award", "World Soundtrack Award", "Gotham Award", "Saturn Award for Science Fiction"], "LOCATION": ["New York"], "MUSICAL_ARTIST": ["Shore"], "COUNTRY": [], "ORGANISATION": ["National Board of Review", "Broadcast Film Critics Association", "Chicago Film Critics"], "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: AWARD, LOCATION, MUSICAL_ARTIST, COUNTRY, ORGANISATION, PERSON\n\nText:\nShore has also been honored with awards from National Board of Review, Recording Academy Honors, Broadcast Film Critics Association, Chicago Film Critics, Genie Award, World Soundtrack Award, New York's Gotham Award, and The Saturn Award for Science Fiction.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, MUSICAL_ARTIST, COUNTRY, ORGANISATION, PERSON\n\nText:\nShore has also been honored with awards from National Board of Review, Recording Academy Honors, Broadcast Film Critics Association, Chicago Film Critics, Genie Award, World Soundtrack Award, New York's Gotham Award, and The Saturn Award for Science Fiction.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_319", "text": "Newspaper sources are not clear whether Chen attempted to enter and was rejected by the Miss World pageant or the Miss Universe pageant.", "requested_types": ["PERSON", "EVENT", "COUNTRY", "MUSIC_GENRE"], "distractors": ["MUSIC_GENRE", "COUNTRY"], "gold": {"PERSON": ["Chen"], "EVENT": ["Miss World pageant", "Miss Universe pageant"], "COUNTRY": [], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, MUSIC_GENRE\n\nText:\nNewspaper sources are not clear whether Chen attempted to enter and was rejected by the Miss World pageant or the Miss Universe pageant.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, MUSIC_GENRE\n\nText:\nNewspaper sources are not clear whether Chen attempted to enter and was rejected by the Miss World pageant or the Miss Universe pageant.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_320", "text": "In Scotland, The Corries, Silly Wizard, Capercaillie, Runrig, Jackie Leven, Julie Fowlis, Karine Polwart, Alasdair Roberts, Dick Gaughan, Wolfstone, Boys of the Lough, and The Silencers have kept Scottish folk vibrant and fresh by mixing traditional Scottish and Gaelic folk songs with more contemporary genres.", "requested_types": ["MUSIC_GENRE", "ALBUM", "BAND", "MUSICAL_ARTIST", "COUNTRY", "AWARD"], "distractors": ["AWARD", "ALBUM"], "gold": {"MUSIC_GENRE": ["Scottish folk", "traditional Scottish", "Gaelic folk"], "ALBUM": [], "BAND": ["The Corries", "Silly Wizard", "Capercaillie", "Runrig", "Wolfstone", "Boys of the Lough", "The Silencers"], "MUSICAL_ARTIST": ["Jackie Leven", "Julie Fowlis", "Karine Polwart", "Alasdair Roberts", "Dick Gaughan"], "COUNTRY": ["Scotland"], "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: MUSIC_GENRE, ALBUM, BAND, MUSICAL_ARTIST, COUNTRY, AWARD\n\nText:\nIn Scotland, The Corries, Silly Wizard, Capercaillie, Runrig, Jackie Leven, Julie Fowlis, Karine Polwart, Alasdair Roberts, Dick Gaughan, Wolfstone, Boys of the Lough, and The Silencers have kept Scottish folk vibrant and fresh by mixing traditional Scottish and Gaelic folk songs with more contemporary genres.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, ALBUM, BAND, MUSICAL_ARTIST, COUNTRY, AWARD\n\nText:\nIn Scotland, The Corries, Silly Wizard, Capercaillie, Runrig, Jackie Leven, Julie Fowlis, Karine Polwart, Alasdair Roberts, Dick Gaughan, Wolfstone, Boys of the Lough, and The Silencers have kept Scottish folk vibrant and fresh by mixing traditional Scottish and Gaelic folk songs with more contemporary genres.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_321", "text": "In conjunction with the book and the play that also paid tribute to his uncle, Milt Gabler, Crystal produced two CD compilations: Billy Crystal Presents: The Milt Gabler Story, which featured his uncle's most influential recordings from Billie Holiday 's Strange Fruit to Rock Around the Clock by Bill Haley & His Comets; and Billy Remembers Billie featuring Crystal's favorite Holiday recordings.", "requested_types": ["BAND", "ALBUM", "MISC", "MUSICAL_INSTRUMENT", "SONG", "PERSON", "MUSICAL_ARTIST", "COUNTRY", "EVENT"], "distractors": ["COUNTRY", "MUSICAL_INSTRUMENT", "EVENT"], "gold": {"BAND": ["Bill Haley & His Comets"], "ALBUM": ["The Milt Gabler Story", "Billy Remembers Billie"], "MISC": ["CD"], "MUSICAL_INSTRUMENT": [], "SONG": ["Strange Fruit", "Rock Around the Clock"], "PERSON": ["Milt Gabler"], "MUSICAL_ARTIST": ["Crystal", "Billy Crystal", "Billie Holiday", "Holiday"], "COUNTRY": [], "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: BAND, ALBUM, MISC, MUSICAL_INSTRUMENT, SONG, PERSON, MUSICAL_ARTIST, COUNTRY, EVENT\n\nText:\nIn conjunction with the book and the play that also paid tribute to his uncle, Milt Gabler, Crystal produced two CD compilations: Billy Crystal Presents: The Milt Gabler Story, which featured his uncle's most influential recordings from Billie Holiday 's Strange Fruit to Rock Around the Clock by Bill Haley & His Comets; and Billy Remembers Billie featuring Crystal's favorite Holiday recordings.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, ALBUM, MISC, MUSICAL_INSTRUMENT, SONG, PERSON, MUSICAL_ARTIST, COUNTRY, EVENT\n\nText:\nIn conjunction with the book and the play that also paid tribute to his uncle, Milt Gabler, Crystal produced two CD compilations: Billy Crystal Presents: The Milt Gabler Story, which featured his uncle's most influential recordings from Billie Holiday 's Strange Fruit to Rock Around the Clock by Bill Haley & His Comets; and Billy Remembers Billie featuring Crystal's favorite Holiday recordings.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_322", "text": "She has won several awards throughout her career, including one Grammy Award from seven nominations, one Latin Grammy Award, ten Juno Awards, one BRIT Award, one Billboard Music Award, one MTV Europe Music Award, one World Music Award, and three Much Music Video Awards.", "requested_types": ["COUNTRY", "AWARD"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "AWARD": ["Grammy Award", "Latin Grammy Award", "Juno Awards", "BRIT Award", "Billboard Music Award", "MTV Europe Music Award", "World Music Award", "Much Music Video Awards"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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\n\nText:\nShe has won several awards throughout her career, including one Grammy Award from seven nominations, one Latin Grammy Award, ten Juno Awards, one BRIT Award, one Billboard Music Award, one MTV Europe Music Award, one World Music Award, and three Much Music Video 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: COUNTRY, AWARD\n\nText:\nShe has won several awards throughout her career, including one Grammy Award from seven nominations, one Latin Grammy Award, ten Juno Awards, one BRIT Award, one Billboard Music Award, one MTV Europe Music Award, one World Music Award, and three Much Music Video 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_music_323", "text": "The awards were established in 1969, and represent a variety of musical styles, including Christian rock, Contemporary Christian music, Christian hip hop, Christian country music, and Urban contemporary gospel.", "requested_types": ["AWARD", "PERSON", "SONG", "MUSIC_GENRE"], "distractors": ["AWARD", "PERSON", "SONG"], "gold": {"AWARD": [], "PERSON": [], "SONG": [], "MUSIC_GENRE": ["Christian rock", "Contemporary Christian music", "Christian hip hop", "Christian country music", "Urban contemporary gospel"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE\n\nText:\nThe awards were established in 1969, and represent a variety of musical styles, including Christian rock, Contemporary Christian music, Christian hip hop, Christian country music, and Urban contemporary gospel.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE\n\nText:\nThe awards were established in 1969, and represent a variety of musical styles, including Christian rock, Contemporary Christian music, Christian hip hop, Christian country music, and Urban contemporary gospel.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_324", "text": "He provided musical direction for the ceremonies of the 1978 Commonwealth Games, EXPO 86, The World University Games, the XV Olympic Winter Games, and for countless television shows.", "requested_types": ["LOCATION", "EVENT", "MISC", "COUNTRY"], "distractors": ["LOCATION", "COUNTRY", "MISC"], "gold": {"LOCATION": [], "EVENT": ["1978 Commonwealth Games", "EXPO 86", "The World University Games", "XV Olympic Winter Games"], "MISC": [], "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: LOCATION, EVENT, MISC, COUNTRY\n\nText:\nHe provided musical direction for the ceremonies of the 1978 Commonwealth Games, EXPO 86, The World University Games, the XV Olympic Winter Games, and for countless television shows.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, COUNTRY\n\nText:\nHe provided musical direction for the ceremonies of the 1978 Commonwealth Games, EXPO 86, The World University Games, the XV Olympic Winter Games, and for countless television shows.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_325", "text": "Chicago blues is influenced to a large extent by the Delta blues style, because many performers had migrated from the Mississippi region.", "requested_types": ["LOCATION", "BAND", "MUSIC_GENRE"], "distractors": ["BAND"], "gold": {"LOCATION": ["Mississippi"], "BAND": [], "MUSIC_GENRE": ["Chicago blues", "Delta blues"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSIC_GENRE\n\nText:\nChicago blues is influenced to a large extent by the Delta blues style, because many performers had migrated from the Mississippi region.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSIC_GENRE\n\nText:\nChicago blues is influenced to a large extent by the Delta blues style, because many performers had migrated from the Mississippi region.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_326", "text": "Performers such as Frank Stokes, Sleepy John Estes, Robert Wilkins, Joe McCoy, Casey Bill Weldon and Memphis Minnie used a variety of unusual instruments such as washboard, fiddle, kazoo or mandolin.", "requested_types": ["MUSIC_GENRE", "MUSICAL_INSTRUMENT", "PERSON", "MUSICAL_ARTIST", "LOCATION"], "distractors": ["MUSIC_GENRE", "PERSON", "LOCATION"], "gold": {"MUSIC_GENRE": [], "MUSICAL_INSTRUMENT": ["washboard", "fiddle", "kazoo", "mandolin"], "PERSON": [], "MUSICAL_ARTIST": ["Frank Stokes", "Sleepy John Estes", "Robert Wilkins", "Joe McCoy", "Casey Bill Weldon", "Memphis Minnie"], "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: MUSIC_GENRE, MUSICAL_INSTRUMENT, PERSON, MUSICAL_ARTIST, LOCATION\n\nText:\nPerformers such as Frank Stokes, Sleepy John Estes, Robert Wilkins, Joe McCoy, Casey Bill Weldon and Memphis Minnie used a variety of unusual instruments such as washboard, fiddle, kazoo or mandolin.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_INSTRUMENT, PERSON, MUSICAL_ARTIST, LOCATION\n\nText:\nPerformers such as Frank Stokes, Sleepy John Estes, Robert Wilkins, Joe McCoy, Casey Bill Weldon and Memphis Minnie used a variety of unusual instruments such as washboard, fiddle, kazoo or mandolin.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_327", "text": "The World DanceSport Federation (WDSF), formerly the International DanceSport Federation (IDSF), is the international governing body of DanceSport and Wheelchair DanceSport, as recognised by the International Olympic Committee (IOC) and the International Paralympic Committee (IPC).", "requested_types": ["BAND", "AWARD", "MISC", "ORGANISATION"], "distractors": ["AWARD", "BAND"], "gold": {"BAND": [], "AWARD": [], "MISC": ["DanceSport", "Wheelchair DanceSport"], "ORGANISATION": ["World DanceSport Federation", "WDSF", "International DanceSport Federation", "IDSF", "International Olympic Committee", "IOC", "International Paralympic Committee", "IPC"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, AWARD, MISC, ORGANISATION\n\nText:\nThe World DanceSport Federation (WDSF), formerly the International DanceSport Federation (IDSF), is the international governing body of DanceSport and Wheelchair DanceSport, as recognised by the International Olympic Committee (IOC) and the International Paralympic Committee (IPC).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, AWARD, MISC, ORGANISATION\n\nText:\nThe World DanceSport Federation (WDSF), formerly the International DanceSport Federation (IDSF), is the international governing body of DanceSport and Wheelchair DanceSport, as recognised by the International Olympic Committee (IOC) and the International Paralympic Committee (IPC).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_328", "text": "While he was working at Friends, Bubbles formed an association with Hawkwind and became responsible for a run of their album sleeves, including In Search of Space, Doremi Fasol Latido and Space Ritual.", "requested_types": ["PERSON", "LOCATION", "ALBUM", "BAND", "SONG", "COUNTRY"], "distractors": ["LOCATION", "COUNTRY", "SONG"], "gold": {"PERSON": ["Bubbles"], "LOCATION": [], "ALBUM": ["In Search of Space", "Doremi Fasol Latido", "Space Ritual"], "BAND": ["Hawkwind"], "SONG": [], "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: PERSON, LOCATION, ALBUM, BAND, SONG, COUNTRY\n\nText:\nWhile he was working at Friends, Bubbles formed an association with Hawkwind and became responsible for a run of their album sleeves, including In Search of Space, Doremi Fasol Latido and Space Ritual.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, BAND, SONG, COUNTRY\n\nText:\nWhile he was working at Friends, Bubbles formed an association with Hawkwind and became responsible for a run of their album sleeves, including In Search of Space, Doremi Fasol Latido and Space Ritual.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_329", "text": "In the 1990s, artists like Me'shell Ndegeocello, Brooklyn Funk Essentials and the (predominantly UK-based) acid jazz movement including artists and bands such as Jamiroquai, Incognito, Galliano, Omar Lye-Fook, Los Tetas and the Brand New Heavies carried on with strong elements of funk.", "requested_types": ["MUSICAL_ARTIST", "MISC", "BAND", "MUSIC_GENRE"], "distractors": ["MISC"], "gold": {"MUSICAL_ARTIST": ["Me'shell Ndegeocello", "Omar Lye-Fook"], "MISC": [], "BAND": ["Brooklyn Funk Essentials", "Jamiroquai", "Incognito", "Galliano", "Los Tetas", "Brand New Heavies"], "MUSIC_GENRE": ["acid jazz", "funk"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MISC, BAND, MUSIC_GENRE\n\nText:\nIn the 1990s, artists like Me'shell Ndegeocello, Brooklyn Funk Essentials and the (predominantly UK-based) acid jazz movement including artists and bands such as Jamiroquai, Incognito, Galliano, Omar Lye-Fook, Los Tetas and the Brand New Heavies carried on with strong elements of funk.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MISC, BAND, MUSIC_GENRE\n\nText:\nIn the 1990s, artists like Me'shell Ndegeocello, Brooklyn Funk Essentials and the (predominantly UK-based) acid jazz movement including artists and bands such as Jamiroquai, Incognito, Galliano, Omar Lye-Fook, Los Tetas and the Brand New Heavies carried on with strong elements of funk.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_330", "text": "The Simpsons episode Homerpalooza features Homer explaining 1970s rock music to Lisa, Bart, and Milhouse: Grand Funk Railroad paved the way for Jefferson Airplane, clearing the way for Jefferson Starship.", "requested_types": ["PERSON", "MISC", "EVENT", "BAND", "COUNTRY", "MUSICAL_ARTIST", "MUSIC_GENRE"], "distractors": ["MUSICAL_ARTIST", "COUNTRY", "EVENT"], "gold": {"PERSON": ["Homerpalooza", "Homer", "Lisa", "Bart", "Milhouse"], "MISC": ["The Simpsons"], "EVENT": [], "BAND": ["Grand Funk Railroad", "Jefferson Airplane", "Jefferson Starship"], "COUNTRY": [], "MUSICAL_ARTIST": [], "MUSIC_GENRE": ["rock music"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT, BAND, COUNTRY, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nThe Simpsons episode Homerpalooza features Homer explaining 1970s rock music to Lisa, Bart, and Milhouse: Grand Funk Railroad paved the way for Jefferson Airplane, clearing the way for Jefferson Starship.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, EVENT, BAND, COUNTRY, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nThe Simpsons episode Homerpalooza features Homer explaining 1970s rock music to Lisa, Bart, and Milhouse: Grand Funk Railroad paved the way for Jefferson Airplane, clearing the way for Jefferson Starship.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_331", "text": "His film Annie Hall (1977), a romantic comedy featuring Allen and his frequent collaborator Diane Keaton, won four Academy Awards, including Academy Award for Best Picture, Academy Award for Best Director, Academy Award for Best Original Screenplay, and Academy Award for Best Actress for Keaton.", "requested_types": ["ALBUM", "PERSON", "AWARD", "MISC"], "distractors": ["ALBUM"], "gold": {"ALBUM": [], "PERSON": ["Allen", "Diane Keaton", "Keaton"], "AWARD": ["Academy Awards", "Academy Award for Best Picture", "Academy Award for Best Director", "Academy Award for Best Original Screenplay", "Academy Award for Best Actress"], "MISC": ["Annie Hall"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, PERSON, AWARD, MISC\n\nText:\nHis film Annie Hall (1977), a romantic comedy featuring Allen and his frequent collaborator Diane Keaton, won four Academy Awards, including Academy Award for Best Picture, Academy Award for Best Director, Academy Award for Best Original Screenplay, and Academy Award for Best Actress for Keaton.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, PERSON, AWARD, MISC\n\nText:\nHis film Annie Hall (1977), a romantic comedy featuring Allen and his frequent collaborator Diane Keaton, won four Academy Awards, including Academy Award for Best Picture, Academy Award for Best Director, Academy Award for Best Original Screenplay, and Academy Award for Best Actress for Keaton.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_332", "text": "Armstrong had nineteen Top Ten records including Stardust, What a Wonderful World, When The Saints Go Marching In, Dream a Little Dream of Me, Ain't Misbehavin ', You Rascal You, and Stompin' at the Savoy.", "requested_types": ["ALBUM", "SONG", "EVENT", "MUSICAL_ARTIST", "MUSIC_GENRE"], "distractors": ["MUSIC_GENRE", "ALBUM", "EVENT"], "gold": {"ALBUM": [], "SONG": ["Stardust", "What a Wonderful World", "When The Saints Go Marching In", "Dream a Little Dream of Me", "Ain't Misbehavin '", "You Rascal You", "Stompin 'at the Savoy"], "EVENT": [], "MUSICAL_ARTIST": ["Armstrong"], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, SONG, EVENT, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nArmstrong had nineteen Top Ten records including Stardust, What a Wonderful World, When The Saints Go Marching In, Dream a Little Dream of Me, Ain't Misbehavin ', You Rascal You, and Stompin' at the Savoy.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, SONG, EVENT, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nArmstrong had nineteen Top Ten records including Stardust, What a Wonderful World, When The Saints Go Marching In, Dream a Little Dream of Me, Ain't Misbehavin ', You Rascal You, and Stompin' at the Savoy.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_333", "text": "It won five Grammys: Grammy Award for Best Rock Album, Grammy Award for Best Rock Song (Dani California), Grammy Award for Best Rock Performance by a Duo or Group with Vocal (Dani California), Best Boxed Or Special Limited Edition Package, and Best Producer (Rick Rubin).", "requested_types": ["PERSON", "AWARD", "ALBUM", "SONG"], "distractors": ["ALBUM", "PERSON"], "gold": {"PERSON": [], "AWARD": ["Grammy Award for Best Rock Album", "Grammy Award for Best Rock Song", "Grammy Award for Best Rock Performance by a Duo or Group with Vocal", "Best Boxed Or Special Limited Edition Package", "Best Producer"], "ALBUM": [], "SONG": ["Dani California", "Rick Rubin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, SONG\n\nText:\nIt won five Grammys: Grammy Award for Best Rock Album, Grammy Award for Best Rock Song (Dani California), Grammy Award for Best Rock Performance by a Duo or Group with Vocal (Dani California), Best Boxed Or Special Limited Edition Package, and Best Producer (Rick Rubin).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, SONG\n\nText:\nIt won five Grammys: Grammy Award for Best Rock Album, Grammy Award for Best Rock Song (Dani California), Grammy Award for Best Rock Performance by a Duo or Group with Vocal (Dani California), Best Boxed Or Special Limited Edition Package, and Best Producer (Rick Rubin).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_334", "text": "Muse released another live DVD on 12 December 2005, Absolution Tour, containing edited and remastered highlights from their Glastonbury performance unseen footage from their performances at London Earls Court Exhibition Centre, Wembley Arena, and the Wiltern Theatre in Los Angeles.", "requested_types": ["ALBUM", "EVENT", "COUNTRY", "BAND", "MUSICAL_ARTIST", "LOCATION", "MISC"], "distractors": ["MUSICAL_ARTIST", "MISC", "COUNTRY"], "gold": {"ALBUM": ["Absolution Tour"], "EVENT": ["Glastonbury"], "COUNTRY": [], "BAND": ["Muse"], "MUSICAL_ARTIST": [], "LOCATION": ["London", "Earls Court Exhibition Centre", "Wembley Arena", "Wiltern Theatre", "Los Angeles"], "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: ALBUM, EVENT, COUNTRY, BAND, MUSICAL_ARTIST, LOCATION, MISC\n\nText:\nMuse released another live DVD on 12 December 2005, Absolution Tour, containing edited and remastered highlights from their Glastonbury performance unseen footage from their performances at London Earls Court Exhibition Centre, Wembley Arena, and the Wiltern Theatre in Los Angeles.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, EVENT, COUNTRY, BAND, MUSICAL_ARTIST, LOCATION, MISC\n\nText:\nMuse released another live DVD on 12 December 2005, Absolution Tour, containing edited and remastered highlights from their Glastonbury performance unseen footage from their performances at London Earls Court Exhibition Centre, Wembley Arena, and the Wiltern Theatre in Los Angeles.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_335", "text": "The band released four albums: Styx I (1972), Styx II (1973), The Serpent Is Rising (1973), and Man of Miracles (1974).", "requested_types": ["AWARD", "MUSICAL_INSTRUMENT", "ALBUM"], "distractors": ["MUSICAL_INSTRUMENT", "AWARD"], "gold": {"AWARD": [], "MUSICAL_INSTRUMENT": [], "ALBUM": ["Styx I", "Styx II", "The Serpent Is Rising", "Man of Miracles"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, ALBUM\n\nText:\nThe band released four albums: Styx I (1972), Styx II (1973), The Serpent Is Rising (1973), and Man of Miracles (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: AWARD, MUSICAL_INSTRUMENT, ALBUM\n\nText:\nThe band released four albums: Styx I (1972), Styx II (1973), The Serpent Is Rising (1973), and Man of Miracles (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_music_336", "text": "Nirvana occasionally played cover songs by these bands, including Led Zeppelin's Heartbreaker, Moby Dick and Immigrant Song, Black Sabbath's Hand of Doom, and Kiss' Do You Love Me? and wrote the Incesticide song Aero Zeppelin as a tribute to Led Zeppelin and Aerosmith.", "requested_types": ["SONG", "MUSICAL_ARTIST", "BAND", "ALBUM", "MISC"], "distractors": ["MISC"], "gold": {"SONG": ["Heartbreaker", "Moby Dick", "Immigrant Song", "Hand of Doom", "Kiss' Do You Love Me?", "Aero Zeppelin"], "MUSICAL_ARTIST": ["Nirvana"], "BAND": ["Led Zeppelin", "Black Sabbath", "Aerosmith"], "ALBUM": ["Incesticide"], "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: SONG, MUSICAL_ARTIST, BAND, ALBUM, MISC\n\nText:\nNirvana occasionally played cover songs by these bands, including Led Zeppelin's Heartbreaker, Moby Dick and Immigrant Song, Black Sabbath's Hand of Doom, and Kiss' Do You Love Me? and wrote the Incesticide song Aero Zeppelin as a tribute to Led Zeppelin and Aerosmith.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, BAND, ALBUM, MISC\n\nText:\nNirvana occasionally played cover songs by these bands, including Led Zeppelin's Heartbreaker, Moby Dick and Immigrant Song, Black Sabbath's Hand of Doom, and Kiss' Do You Love Me? and wrote the Incesticide song Aero Zeppelin as a tribute to Led Zeppelin and Aerosmith.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_337", "text": "It ranges from the combination of live instrumentation with the beats of jazz house (as exemplified by St Germain, Jazzanova, and Fila Brazillia) to more band-based improvised jazz with electronic elements (for example, The Cinematic Orchestra, Kobol and the Norwegian future jazz style pioneered by Bugge Wesseltoft, Jaga Jazzist, and Nils Petter Molvær).", "requested_types": ["MUSICAL_ARTIST", "MISC", "BAND", "LOCATION", "MUSIC_GENRE", "SONG"], "distractors": ["LOCATION", "SONG"], "gold": {"MUSICAL_ARTIST": ["St Germain", "Bugge Wesseltoft", "Nils Petter Molvær"], "MISC": ["Norwegian"], "BAND": ["Jazzanova", "Fila Brazillia", "The Cinematic Orchestra", "Kobol", "Jaga Jazzist"], "LOCATION": [], "MUSIC_GENRE": ["jazz house", "jazz"], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MISC, BAND, LOCATION, MUSIC_GENRE, SONG\n\nText:\nIt ranges from the combination of live instrumentation with the beats of jazz house (as exemplified by St Germain, Jazzanova, and Fila Brazillia) to more band-based improvised jazz with electronic elements (for example, The Cinematic Orchestra, Kobol and the Norwegian future jazz style pioneered by Bugge Wesseltoft, Jaga Jazzist, and Nils Petter Molvær).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MISC, BAND, LOCATION, MUSIC_GENRE, SONG\n\nText:\nIt ranges from the combination of live instrumentation with the beats of jazz house (as exemplified by St Germain, Jazzanova, and Fila Brazillia) to more band-based improvised jazz with electronic elements (for example, The Cinematic Orchestra, Kobol and the Norwegian future jazz style pioneered by Bugge Wesseltoft, Jaga Jazzist, and Nils Petter Molvær).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_338", "text": "Os Mutantes (meaning the mutants;) are an influential Brazilian Rock music band that were linked with the Tropicália movement, a dissident musical movement during the Brazilian dictatorship of the late 1960s.", "requested_types": ["ORGANISATION", "MUSICAL_ARTIST", "MISC", "BAND", "MUSIC_GENRE", "AWARD"], "distractors": ["ORGANISATION", "AWARD", "MUSICAL_ARTIST"], "gold": {"ORGANISATION": [], "MUSICAL_ARTIST": [], "MISC": ["Brazilian"], "BAND": ["Os Mutantes"], "MUSIC_GENRE": ["Rock music", "Tropicália"], "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, MUSICAL_ARTIST, MISC, BAND, MUSIC_GENRE, AWARD\n\nText:\nOs Mutantes (meaning the mutants;) are an influential Brazilian Rock music band that were linked with the Tropicália movement, a dissident musical movement during the Brazilian dictatorship of the late 1960s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, MISC, BAND, MUSIC_GENRE, AWARD\n\nText:\nOs Mutantes (meaning the mutants;) are an influential Brazilian Rock music band that were linked with the Tropicália movement, a dissident musical movement during the Brazilian dictatorship of the late 1960s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_339", "text": "Distinct from Afrobeat is Afrobeats - a sound originating in West Africa in the 21st century, one which takes in diverse influences and is an eclectic combination of genres such as British house music, hiplife, hip hop, dancehall, Soca music, Jùjú music, highlife, Rhythm and blues, Ndombolo, Naija beats, Azonto, and Palm-wine music.", "requested_types": ["LOCATION", "MUSIC_GENRE", "MUSICAL_INSTRUMENT", "MISC", "COUNTRY"], "distractors": ["MUSICAL_INSTRUMENT", "COUNTRY"], "gold": {"LOCATION": ["West Africa"], "MUSIC_GENRE": ["Afrobeat", "Afrobeats", "house music", "hiplife", "hip hop", "dancehall", "Soca music", "Jùjú music", "highlife", "Rhythm and blues", "Ndombolo", "Naija beats", "Azonto", "Palm-wine music"], "MUSICAL_INSTRUMENT": [], "MISC": ["British"], "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: LOCATION, MUSIC_GENRE, MUSICAL_INSTRUMENT, MISC, COUNTRY\n\nText:\nDistinct from Afrobeat is Afrobeats - a sound originating in West Africa in the 21st century, one which takes in diverse influences and is an eclectic combination of genres such as British house music, hiplife, hip hop, dancehall, Soca music, Jùjú music, highlife, Rhythm and blues, Ndombolo, Naija beats, Azonto, and Palm-wine music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, MUSICAL_INSTRUMENT, MISC, COUNTRY\n\nText:\nDistinct from Afrobeat is Afrobeats - a sound originating in West Africa in the 21st century, one which takes in diverse influences and is an eclectic combination of genres such as British house music, hiplife, hip hop, dancehall, Soca music, Jùjú music, highlife, Rhythm and blues, Ndombolo, Naija beats, Azonto, and Palm-wine music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_340", "text": "Priories may be houses of mendicant friar s or nun s (such as the Dominican Order, Augustinians, Franciscans, and Carmelites, for instance), or monasteries of monk s or nun s (as with the Benedictines).", "requested_types": ["ORGANISATION", "BAND"], "distractors": ["BAND"], "gold": {"ORGANISATION": ["Dominican Order", "Augustinians", "Franciscans", "Carmelites", "Benedictines"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nPriories may be houses of mendicant friar s or nun s (such as the Dominican Order, Augustinians, Franciscans, and Carmelites, for instance), or monasteries of monk s or nun s (as with the Benedictines).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nPriories may be houses of mendicant friar s or nun s (such as the Dominican Order, Augustinians, Franciscans, and Carmelites, for instance), or monasteries of monk s or nun s (as with the Benedictines).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_341", "text": "Along with one other Original Six indoor ice hockey arena, the Boston Garden, the Montreal Forum used a high-pitched siren to signal the end of an NHL game's period - the siren would later be re-installed in the Forum's successor facility, the Bell Centre (and still in use there), much as the TD Garden in Boston inherited the lower-pitched Garden's siren.", "requested_types": ["LOCATION", "MISC", "ALBUM"], "distractors": ["ALBUM"], "gold": {"LOCATION": ["Boston Garden", "Montreal Forum", "Forum", "Bell Centre", "TD Garden", "Boston", "Garden"], "MISC": ["NHL game"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nAlong with one other Original Six indoor ice hockey arena, the Boston Garden, the Montreal Forum used a high-pitched siren to signal the end of an NHL game's period - the siren would later be re-installed in the Forum's successor facility, the Bell Centre (and still in use there), much as the TD Garden in Boston inherited the lower-pitched Garden's siren.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nAlong with one other Original Six indoor ice hockey arena, the Boston Garden, the Montreal Forum used a high-pitched siren to signal the end of an NHL game's period - the siren would later be re-installed in the Forum's successor facility, the Bell Centre (and still in use there), much as the TD Garden in Boston inherited the lower-pitched Garden's siren.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_342", "text": "At the 2014 Brit Awards on 19 February, Bowie became the oldest recipient of a Brit Awards in the ceremony's history when he won the award for Brit Award for British Male Solo Artist, which was collected on his behalf by Kate Moss.", "requested_types": ["ORGANISATION", "MUSICAL_INSTRUMENT", "AWARD", "MUSICAL_ARTIST", "PERSON"], "distractors": ["ORGANISATION", "MUSICAL_INSTRUMENT"], "gold": {"ORGANISATION": [], "MUSICAL_INSTRUMENT": [], "AWARD": ["2014 Brit Awards", "Brit Awards", "Brit Award for British Male Solo Artist"], "MUSICAL_ARTIST": ["Bowie"], "PERSON": ["Kate Moss"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, AWARD, MUSICAL_ARTIST, PERSON\n\nText:\nAt the 2014 Brit Awards on 19 February, Bowie became the oldest recipient of a Brit Awards in the ceremony's history when he won the award for Brit Award for British Male Solo Artist, which was collected on his behalf by Kate Moss.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, AWARD, MUSICAL_ARTIST, PERSON\n\nText:\nAt the 2014 Brit Awards on 19 February, Bowie became the oldest recipient of a Brit Awards in the ceremony's history when he won the award for Brit Award for British Male Solo Artist, which was collected on his behalf by Kate Moss.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_343", "text": "In Southeast Asia, local boy bands also emerged as a result of the continued success of Korean and Japanese boy bands such as SMAP, Shinhwa, TVXQ!, Arashi, Exile, Super Junior, Big Bang, SHINee, EXO, and BTS.", "requested_types": ["LOCATION", "PERSON", "BAND", "SONG", "AWARD", "MISC"], "distractors": ["PERSON", "SONG", "AWARD"], "gold": {"LOCATION": ["Southeast Asia"], "PERSON": [], "BAND": ["SMAP", "Shinhwa", "TVXQ!", "Arashi", "Exile", "Super Junior", "Big Bang", "SHINee", "EXO", "BTS"], "SONG": [], "AWARD": [], "MISC": ["Korean", "Japanese"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, SONG, AWARD, MISC\n\nText:\nIn Southeast Asia, local boy bands also emerged as a result of the continued success of Korean and Japanese boy bands such as SMAP, Shinhwa, TVXQ!, Arashi, Exile, Super Junior, Big Bang, SHINee, EXO, and BTS.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, SONG, AWARD, MISC\n\nText:\nIn Southeast Asia, local boy bands also emerged as a result of the continued success of Korean and Japanese boy bands such as SMAP, Shinhwa, TVXQ!, Arashi, Exile, Super Junior, Big Bang, SHINee, EXO, and BTS.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_344", "text": "He began a long-running DJ residency at London's Met Bar in 2003, playing a mix of House music and Hip hop music, and has gone on to become a renowned DJ over recent years, performing sets at Cielo New York and Pacha Ibiza.", "requested_types": ["LOCATION", "MUSIC_GENRE", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"LOCATION": ["London's Met Bar", "Cielo New York", "Pacha Ibiza"], "MUSIC_GENRE": ["House music", "Hip hop music"], "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: LOCATION, MUSIC_GENRE, COUNTRY\n\nText:\nHe began a long-running DJ residency at London's Met Bar in 2003, playing a mix of House music and Hip hop music, and has gone on to become a renowned DJ over recent years, performing sets at Cielo New York and Pacha Ibiza.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, COUNTRY\n\nText:\nHe began a long-running DJ residency at London's Met Bar in 2003, playing a mix of House music and Hip hop music, and has gone on to become a renowned DJ over recent years, performing sets at Cielo New York and Pacha Ibiza.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_345", "text": "Other artists such as Brooks and Dunn (Boot Scootin 'Boogie) also combined conventional country with slick, rock elements, while Lorrie Morgan, Mary Chapin Carpenter, and Kathy Mattea updated neotraditionalist styles.", "requested_types": ["SONG", "MUSIC_GENRE", "MUSICAL_ARTIST", "BAND", "MISC", "COUNTRY"], "distractors": ["COUNTRY", "MISC"], "gold": {"SONG": ["Boot Scootin 'Boogie"], "MUSIC_GENRE": ["country", "rock"], "MUSICAL_ARTIST": ["Lorrie Morgan", "Mary Chapin Carpenter", "Kathy Mattea"], "BAND": ["Brooks and Dunn"], "MISC": [], "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: SONG, MUSIC_GENRE, MUSICAL_ARTIST, BAND, MISC, COUNTRY\n\nText:\nOther artists such as Brooks and Dunn (Boot Scootin 'Boogie) also combined conventional country with slick, rock elements, while Lorrie Morgan, Mary Chapin Carpenter, and Kathy Mattea updated neotraditionalist styles.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE, MUSICAL_ARTIST, BAND, MISC, COUNTRY\n\nText:\nOther artists such as Brooks and Dunn (Boot Scootin 'Boogie) also combined conventional country with slick, rock elements, while Lorrie Morgan, Mary Chapin Carpenter, and Kathy Mattea updated neotraditionalist styles.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_346", "text": "About the musical style of the album Pitchfork writer Joe Tangari wrote that Generation Terrorists walked a weird line between agit- Punk rock, cock rock, romantic melodicism and Glam rock, and was so obviously patterned after the Clash's London Calling that it was actually kind of cute.", "requested_types": ["BAND", "MUSIC_GENRE", "ORGANISATION", "ALBUM", "LOCATION", "PERSON"], "distractors": ["LOCATION"], "gold": {"BAND": ["Clash"], "MUSIC_GENRE": ["Punk rock", "cock rock", "romantic melodicism", "Glam rock"], "ORGANISATION": ["Pitchfork"], "ALBUM": ["Generation Terrorists", "London Calling"], "LOCATION": [], "PERSON": ["Joe Tangari"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, ORGANISATION, ALBUM, LOCATION, PERSON\n\nText:\nAbout the musical style of the album Pitchfork writer Joe Tangari wrote that Generation Terrorists walked a weird line between agit- Punk rock, cock rock, romantic melodicism and Glam rock, and was so obviously patterned after the Clash's London Calling that it was actually kind of cute.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, ORGANISATION, ALBUM, LOCATION, PERSON\n\nText:\nAbout the musical style of the album Pitchfork writer Joe Tangari wrote that Generation Terrorists walked a weird line between agit- Punk rock, cock rock, romantic melodicism and Glam rock, and was so obviously patterned after the Clash's London Calling that it was actually kind of cute.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_347", "text": "At the ARIA Music Awards of 1995 Arena was nominated in six categories and won four trophies: ARIA Award for Best Pop Release and Song of the Year for Chains; and ARIA Award for Album of the Year and ARIA Award for Best Female Artist for Don't Ask. ARIA Music Awards for Tina Arena:", "requested_types": ["ORGANISATION", "MUSICAL_ARTIST", "BAND", "AWARD", "SONG", "LOCATION", "ALBUM"], "distractors": ["BAND", "LOCATION", "ORGANISATION"], "gold": {"ORGANISATION": [], "MUSICAL_ARTIST": ["Arena", "Tina Arena"], "BAND": [], "AWARD": ["ARIA Music Awards", "ARIA Award for Best Pop Release", "Song of the Year", "ARIA Award for Album of the Year", "ARIA Award for Best Female Artist"], "SONG": ["Chains"], "LOCATION": [], "ALBUM": ["Don't Ask"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, BAND, AWARD, SONG, LOCATION, ALBUM\n\nText:\nAt the ARIA Music Awards of 1995 Arena was nominated in six categories and won four trophies: ARIA Award for Best Pop Release and Song of the Year for Chains; and ARIA Award for Album of the Year and ARIA Award for Best Female Artist for Don't Ask. ARIA Music Awards for Tina Arena:\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, BAND, AWARD, SONG, LOCATION, ALBUM\n\nText:\nAt the ARIA Music Awards of 1995 Arena was nominated in six categories and won four trophies: ARIA Award for Best Pop Release and Song of the Year for Chains; and ARIA Award for Album of the Year and ARIA Award for Best Female Artist for Don't Ask. ARIA Music Awards for Tina Arena:\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_348", "text": "In 2019, Cayabyab composed the theme song for the 2019 Southeast Asian Games, We Win as One; with lyrics by playwright Floy Quintos, and sung by Lea Salonga.", "requested_types": ["LOCATION", "MUSICAL_ARTIST", "SONG", "PERSON", "EVENT", "MUSICAL_INSTRUMENT"], "distractors": ["LOCATION", "MUSICAL_INSTRUMENT"], "gold": {"LOCATION": [], "MUSICAL_ARTIST": ["Cayabyab", "Lea Salonga"], "SONG": ["We Win as One"], "PERSON": ["Floy Quintos"], "EVENT": ["2019 Southeast Asian Games"], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, SONG, PERSON, EVENT, MUSICAL_INSTRUMENT\n\nText:\nIn 2019, Cayabyab composed the theme song for the 2019 Southeast Asian Games, We Win as One; with lyrics by playwright Floy Quintos, and sung by Lea Salonga.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, SONG, PERSON, EVENT, MUSICAL_INSTRUMENT\n\nText:\nIn 2019, Cayabyab composed the theme song for the 2019 Southeast Asian Games, We Win as One; with lyrics by playwright Floy Quintos, and sung by Lea Salonga.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_349", "text": "A 1984 New York Times article on the emerging aesthetic acknowledged cowpunk as one of several catch-all terms critics were using to categorize the Country music -influenced music of otherwise unrelated Punk rock and New wave music bands.", "requested_types": ["MUSIC_GENRE", "SONG", "ORGANISATION"], "distractors": ["SONG"], "gold": {"MUSIC_GENRE": ["cowpunk", "Country music", "Punk rock", "New wave music"], "SONG": [], "ORGANISATION": ["New York Times"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, SONG, ORGANISATION\n\nText:\nA 1984 New York Times article on the emerging aesthetic acknowledged cowpunk as one of several catch-all terms critics were using to categorize the Country music -influenced music of otherwise unrelated Punk rock and New wave music bands.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, SONG, ORGANISATION\n\nText:\nA 1984 New York Times article on the emerging aesthetic acknowledged cowpunk as one of several catch-all terms critics were using to categorize the Country music -influenced music of otherwise unrelated Punk rock and New wave music bands.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_350", "text": "Brooks and Dunn have more Country Music Association awards and Academy of Country Music awards than any act in the history of country music.", "requested_types": ["AWARD", "MUSIC_GENRE", "LOCATION", "BAND"], "distractors": ["LOCATION"], "gold": {"AWARD": ["Country Music Association awards", "Academy of Country Music awards"], "MUSIC_GENRE": ["country"], "LOCATION": [], "BAND": ["Brooks and Dunn"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, LOCATION, BAND\n\nText:\nBrooks and Dunn have more Country Music Association awards and Academy of Country Music awards than any act in the history of country music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, LOCATION, BAND\n\nText:\nBrooks and Dunn have more Country Music Association awards and Academy of Country Music awards than any act in the history of country music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_351", "text": "Byrne has contributed songs to five AIDS benefit compilation albums produced by the Red Hot Organization: Red Hot + Blue: A Tribute to Cole Porter, Red Hot + Rio, Silencio = Muerte: Red Hot + Latin, Onda Sonora: Red Hot + Lisbon, and Offbeat: A Red Hot Soundtrip.", "requested_types": ["ALBUM", "MISC", "COUNTRY", "MUSICAL_ARTIST", "ORGANISATION", "BAND"], "distractors": ["BAND", "COUNTRY"], "gold": {"ALBUM": ["Red Hot + Blue: A Tribute to Cole Porter", "Red Hot + Rio", "Silencio = Muerte: Red Hot + Latin", "Onda Sonora: Red Hot + Lisbon", "Offbeat: A Red Hot Soundtrip"], "MISC": ["AIDS"], "COUNTRY": [], "MUSICAL_ARTIST": ["Byrne"], "ORGANISATION": ["Red Hot Organization"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MISC, COUNTRY, MUSICAL_ARTIST, ORGANISATION, BAND\n\nText:\nByrne has contributed songs to five AIDS benefit compilation albums produced by the Red Hot Organization: Red Hot + Blue: A Tribute to Cole Porter, Red Hot + Rio, Silencio = Muerte: Red Hot + Latin, Onda Sonora: Red Hot + Lisbon, and Offbeat: A Red Hot Soundtrip.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MISC, COUNTRY, MUSICAL_ARTIST, ORGANISATION, BAND\n\nText:\nByrne has contributed songs to five AIDS benefit compilation albums produced by the Red Hot Organization: Red Hot + Blue: A Tribute to Cole Porter, Red Hot + Rio, Silencio = Muerte: Red Hot + Latin, Onda Sonora: Red Hot + Lisbon, and Offbeat: A Red Hot Soundtrip.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_352", "text": "Three UEFA Euro 2012 Group D matches, a quarter-final and the final were scheduled for here (with the other matches in Group D being played at the Donbass Arena).", "requested_types": ["ORGANISATION", "MISC", "EVENT", "LOCATION"], "distractors": ["ORGANISATION", "MISC"], "gold": {"ORGANISATION": [], "MISC": [], "EVENT": ["UEFA Euro 2012 Group D"], "LOCATION": ["Donbass Arena"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION\n\nText:\nThree UEFA Euro 2012 Group D matches, a quarter-final and the final were scheduled for here (with the other matches in Group D being played at the Donbass Arena).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION\n\nText:\nThree UEFA Euro 2012 Group D matches, a quarter-final and the final were scheduled for here (with the other matches in Group D being played at the Donbass Arena).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_353", "text": "He also collected in Moldavia, Wallachia, and (in 1913) Algeria.", "requested_types": ["EVENT", "MISC", "COUNTRY", "LOCATION"], "distractors": ["EVENT", "MISC"], "gold": {"EVENT": [], "MISC": [], "COUNTRY": ["Moldavia", "Algeria"], "LOCATION": ["Wallachia"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, LOCATION\n\nText:\nHe also collected in Moldavia, Wallachia, and (in 1913) Algeria.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, LOCATION\n\nText:\nHe also collected in Moldavia, Wallachia, and (in 1913) Algeria.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_354", "text": "Other influential folk artists include Surinder Shinda - famous for his Putt Jattan De - Harbhajan Mann, Manmohan Waris, Meshi Eshara, Sarbjit Cheema, Hans Raj Hans, Sardool Sikander, Anakhi.", "requested_types": ["PERSON", "MUSICAL_INSTRUMENT", "BAND", "MUSICAL_ARTIST", "MISC", "MUSIC_GENRE"], "distractors": ["MUSICAL_INSTRUMENT", "PERSON"], "gold": {"PERSON": [], "MUSICAL_INSTRUMENT": [], "BAND": ["Anakhi"], "MUSICAL_ARTIST": ["Surinder Shinda", "Harbhajan Mann", "Manmohan Waris", "Meshi Eshara", "Sarbjit Cheema", "Hans Raj Hans", "Sardool Sikander"], "MISC": ["Putt Jattan De"], "MUSIC_GENRE": ["folk"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, BAND, MUSICAL_ARTIST, MISC, MUSIC_GENRE\n\nText:\nOther influential folk artists include Surinder Shinda - famous for his Putt Jattan De - Harbhajan Mann, Manmohan Waris, Meshi Eshara, Sarbjit Cheema, Hans Raj Hans, Sardool Sikander, Anakhi.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, BAND, MUSICAL_ARTIST, MISC, MUSIC_GENRE\n\nText:\nOther influential folk artists include Surinder Shinda - famous for his Putt Jattan De - Harbhajan Mann, Manmohan Waris, Meshi Eshara, Sarbjit Cheema, Hans Raj Hans, Sardool Sikander, Anakhi.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_355", "text": "The film received seven Academy Award nominations, winning the Academy Award for Best Actress Oscar for Spacek, a gold album for the soundtrack album, a Grammy nomination for Spacek, Country Music Association and Academy of Country Music awards, and several Golden Globe awards.", "requested_types": ["MUSICAL_INSTRUMENT", "AWARD", "BAND", "PERSON", "MISC", "ORGANISATION"], "distractors": ["MUSICAL_INSTRUMENT", "BAND"], "gold": {"MUSICAL_INSTRUMENT": [], "AWARD": ["Academy Award", "Academy Award for Best Actress", "Academy of Country Music awards", "Golden Globe awards"], "BAND": [], "PERSON": ["Spacek"], "MISC": ["Grammy nomination"], "ORGANISATION": ["Country Music 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: MUSICAL_INSTRUMENT, AWARD, BAND, PERSON, MISC, ORGANISATION\n\nText:\nThe film received seven Academy Award nominations, winning the Academy Award for Best Actress Oscar for Spacek, a gold album for the soundtrack album, a Grammy nomination for Spacek, Country Music Association and Academy of Country Music awards, and several Golden Globe 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: MUSICAL_INSTRUMENT, AWARD, BAND, PERSON, MISC, ORGANISATION\n\nText:\nThe film received seven Academy Award nominations, winning the Academy Award for Best Actress Oscar for Spacek, a gold album for the soundtrack album, a Grammy nomination for Spacek, Country Music Association and Academy of Country Music awards, and several Golden Globe 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_music_356", "text": "Traditional power metal bands like Sweden's HammerFall, England's DragonForce, and America's Iced Earth have a sound clearly indebted to the classic NWOBHM style.See, e.g., Reesman Bryan.", "requested_types": ["COUNTRY", "BAND", "PERSON", "MUSIC_GENRE", "LOCATION"], "distractors": ["LOCATION"], "gold": {"COUNTRY": ["Sweden", "England", "America"], "BAND": ["HammerFall", "DragonForce", "Iced Earth"], "PERSON": ["Reesman Bryan"], "MUSIC_GENRE": ["power metal", "NWOBHM"], "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: COUNTRY, BAND, PERSON, MUSIC_GENRE, LOCATION\n\nText:\nTraditional power metal bands like Sweden's HammerFall, England's DragonForce, and America's Iced Earth have a sound clearly indebted to the classic NWOBHM style.See, e.g., Reesman Bryan.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, PERSON, MUSIC_GENRE, LOCATION\n\nText:\nTraditional power metal bands like Sweden's HammerFall, England's DragonForce, and America's Iced Earth have a sound clearly indebted to the classic NWOBHM style.See, e.g., Reesman Bryan.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_357", "text": "Other songs from 2015 like I Don't Like It, I Love It by Flo Rida, Adventure of a Lifetime by Coldplay, Back Together by Robin Thicke and Levels by Nick Jonas feature disco elements as well.", "requested_types": ["SONG", "MUSIC_GENRE", "MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "BAND", "COUNTRY"], "distractors": ["COUNTRY", "MUSICAL_INSTRUMENT"], "gold": {"SONG": ["I Don't Like It, I Love It", "Adventure of a Lifetime", "Back Together", "Levels"], "MUSIC_GENRE": ["disco"], "MUSICAL_ARTIST": ["Flo Rida", "Robin Thicke", "Nick Jonas"], "MUSICAL_INSTRUMENT": [], "BAND": ["Coldplay"], "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: SONG, MUSIC_GENRE, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, BAND, COUNTRY\n\nText:\nOther songs from 2015 like I Don't Like It, I Love It by Flo Rida, Adventure of a Lifetime by Coldplay, Back Together by Robin Thicke and Levels by Nick Jonas feature disco elements 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: SONG, MUSIC_GENRE, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, BAND, COUNTRY\n\nText:\nOther songs from 2015 like I Don't Like It, I Love It by Flo Rida, Adventure of a Lifetime by Coldplay, Back Together by Robin Thicke and Levels by Nick Jonas feature disco elements 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_music_358", "text": "The band's biggest hit singles are Sentimental ballad such as Easy, Three Times a Lady, and Nightshift; and funky dance hits which include Brick House, Fancy Dancer, Lady (You Bring Me Up), and Too Hot ta Trot.", "requested_types": ["SONG", "MUSIC_GENRE", "ORGANISATION"], "distractors": ["ORGANISATION"], "gold": {"SONG": ["Easy", "Three Times a Lady", "Nightshift", "Brick House", "Fancy Dancer", "Lady", "You Bring Me Up", "Too Hot ta Trot"], "MUSIC_GENRE": ["Sentimental ballad", "funky"], "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: SONG, MUSIC_GENRE, ORGANISATION\n\nText:\nThe band's biggest hit singles are Sentimental ballad such as Easy, Three Times a Lady, and Nightshift; and funky dance hits which include Brick House, Fancy Dancer, Lady (You Bring Me Up), and Too Hot ta Trot.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE, ORGANISATION\n\nText:\nThe band's biggest hit singles are Sentimental ballad such as Easy, Three Times a Lady, and Nightshift; and funky dance hits which include Brick House, Fancy Dancer, Lady (You Bring Me Up), and Too Hot ta Trot.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_359", "text": "More recently, Illinois Jacquet, Ray Pizzi, Frank Tiberi, and Marshall Allen have both doubled on bassoon in addition to their saxophone performances.", "requested_types": ["BAND", "MUSICAL_ARTIST", "EVENT", "MUSICAL_INSTRUMENT", "PERSON"], "distractors": ["PERSON", "EVENT", "BAND"], "gold": {"BAND": [], "MUSICAL_ARTIST": ["Illinois Jacquet", "Ray Pizzi", "Frank Tiberi", "Marshall Allen"], "EVENT": [], "MUSICAL_INSTRUMENT": ["bassoon", "saxophone"], "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: BAND, MUSICAL_ARTIST, EVENT, MUSICAL_INSTRUMENT, PERSON\n\nText:\nMore recently, Illinois Jacquet, Ray Pizzi, Frank Tiberi, and Marshall Allen have both doubled on bassoon in addition to their saxophone performances.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSICAL_ARTIST, EVENT, MUSICAL_INSTRUMENT, PERSON\n\nText:\nMore recently, Illinois Jacquet, Ray Pizzi, Frank Tiberi, and Marshall Allen have both doubled on bassoon in addition to their saxophone performances.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_360", "text": "Influenced by 1960s rhythm and blues, Soul music, Bluegrass music, and Rock music bands such as the Byrds and Buffalo Springfield,", "requested_types": ["SONG", "COUNTRY", "BAND", "MUSIC_GENRE"], "distractors": ["COUNTRY", "SONG"], "gold": {"SONG": [], "COUNTRY": [], "BAND": ["the Byrds", "Buffalo Springfield"], "MUSIC_GENRE": ["rhythm and blues", "Soul music", "Bluegrass music", "Rock music"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, COUNTRY, BAND, MUSIC_GENRE\n\nText:\nInfluenced by 1960s rhythm and blues, Soul music, Bluegrass music, and Rock music bands such as the Byrds and Buffalo Springfield,\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, COUNTRY, BAND, MUSIC_GENRE\n\nText:\nInfluenced by 1960s rhythm and blues, Soul music, Bluegrass music, and Rock music bands such as the Byrds and Buffalo Springfield,\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_361", "text": "Hailed as a great innovator of rock and roll by contemporaries such as Little Richard and Johnny Otis, He is also inducted into the St. Louis Walk of Fame, the Clarksdale Walk of Fame, the Mississippi Musicians Hall of Fame, the Blues Hall of Fame, and the Rhythm and Blues Music Hall of Fame.", "requested_types": ["MUSIC_GENRE", "MUSICAL_ARTIST", "AWARD", "BAND"], "distractors": ["BAND"], "gold": {"MUSIC_GENRE": ["rock and roll"], "MUSICAL_ARTIST": ["Hailed", "Little Richard", "Johnny Otis"], "AWARD": ["St. Louis Walk of Fame", "Clarksdale Walk of Fame", "Mississippi Musicians Hall of Fame", "Blues Hall of Fame", "Rhythm and Blues Music Hall of Fame"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, AWARD, BAND\n\nText:\nHailed as a great innovator of rock and roll by contemporaries such as Little Richard and Johnny Otis, He is also inducted into the St. Louis Walk of Fame, the Clarksdale Walk of Fame, the Mississippi Musicians Hall of Fame, the Blues Hall of Fame, and the Rhythm and Blues Music Hall of Fame.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, AWARD, BAND\n\nText:\nHailed as a great innovator of rock and roll by contemporaries such as Little Richard and Johnny Otis, He is also inducted into the St. Louis Walk of Fame, the Clarksdale Walk of Fame, the Mississippi Musicians Hall of Fame, the Blues Hall of Fame, and the Rhythm and Blues Music Hall of Fame.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_362", "text": "By the 1990s, country music had attained crossover success in the pop charts, with artists like James Blundell and James Reyne singing The Dingoes, and country star Kasey Chambers winning the ARIA Award for Best Female Artist in 2000, 2002 and 2004, tying with pop stars Wendy Matthews and Sia for the most wins in that category.", "requested_types": ["BAND", "AWARD", "ORGANISATION", "MUSICAL_ARTIST", "MUSIC_GENRE"], "distractors": ["ORGANISATION"], "gold": {"BAND": ["The Dingoes"], "AWARD": ["ARIA Award for Best Female Artist"], "ORGANISATION": [], "MUSICAL_ARTIST": ["James Blundell", "James Reyne", "Kasey Chambers", "Wendy Matthews", "Sia"], "MUSIC_GENRE": ["country music", "pop", "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: BAND, AWARD, ORGANISATION, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nBy the 1990s, country music had attained crossover success in the pop charts, with artists like James Blundell and James Reyne singing The Dingoes, and country star Kasey Chambers winning the ARIA Award for Best Female Artist in 2000, 2002 and 2004, tying with pop stars Wendy Matthews and Sia for the most wins in that 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: BAND, AWARD, ORGANISATION, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nBy the 1990s, country music had attained crossover success in the pop charts, with artists like James Blundell and James Reyne singing The Dingoes, and country star Kasey Chambers winning the ARIA Award for Best Female Artist in 2000, 2002 and 2004, tying with pop stars Wendy Matthews and Sia for the most wins in that 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_music_363", "text": "After Ice Cube made a cameo appearance in Dr. Dre's Let Me Ride video in 1993, the two recorded the hit song Natural Born Killaz for Snoop Doggy Dogg 's 1994 short film and soundtrack Murder Was the Case.", "requested_types": ["COUNTRY", "ALBUM", "SONG", "MISC", "PERSON", "MUSICAL_ARTIST"], "distractors": ["MISC", "COUNTRY", "PERSON"], "gold": {"COUNTRY": [], "ALBUM": ["Murder Was the Case"], "SONG": ["Let Me Ride", "Natural Born Killaz"], "MISC": [], "PERSON": [], "MUSICAL_ARTIST": ["Ice Cube", "Dr. Dre", "Snoop Doggy Dogg"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, SONG, MISC, PERSON, MUSICAL_ARTIST\n\nText:\nAfter Ice Cube made a cameo appearance in Dr. Dre's Let Me Ride video in 1993, the two recorded the hit song Natural Born Killaz for Snoop Doggy Dogg 's 1994 short film and soundtrack Murder Was the Case.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, SONG, MISC, PERSON, MUSICAL_ARTIST\n\nText:\nAfter Ice Cube made a cameo appearance in Dr. Dre's Let Me Ride video in 1993, the two recorded the hit song Natural Born Killaz for Snoop Doggy Dogg 's 1994 short film and soundtrack Murder Was the Case.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_364", "text": "The group's original lineup consisted of brothers Brian Wilson, Dennis Wilson, and Carl Wilson, their cousin Mike Love, and their friend Al Jardine.", "requested_types": ["SONG", "MUSICAL_ARTIST", "BAND", "EVENT"], "distractors": ["SONG", "EVENT", "BAND"], "gold": {"SONG": [], "MUSICAL_ARTIST": ["Brian Wilson", "Dennis Wilson", "Carl Wilson", "Mike Love", "Al Jardine"], "BAND": [], "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: SONG, MUSICAL_ARTIST, BAND, EVENT\n\nText:\nThe group's original lineup consisted of brothers Brian Wilson, Dennis Wilson, and Carl Wilson, their cousin Mike Love, and their friend Al Jardine.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, BAND, EVENT\n\nText:\nThe group's original lineup consisted of brothers Brian Wilson, Dennis Wilson, and Carl Wilson, their cousin Mike Love, and their friend Al Jardine.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_365", "text": "The station broadcast Christian feature stories, preaching and teaching to various fundamentalist and Pentecostal denominations and playing Traditional black gospel, Southern gospel, and inspirational music.", "requested_types": ["MUSIC_GENRE", "COUNTRY", "BAND", "MISC"], "distractors": ["BAND", "COUNTRY"], "gold": {"MUSIC_GENRE": ["Traditional black gospel", "Southern gospel"], "COUNTRY": [], "BAND": [], "MISC": ["Christian", "fundamentalist", "Pentecostal denominations"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, COUNTRY, BAND, MISC\n\nText:\nThe station broadcast Christian feature stories, preaching and teaching to various fundamentalist and Pentecostal denominations and playing Traditional black gospel, Southern gospel, and inspirational music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, COUNTRY, BAND, MISC\n\nText:\nThe station broadcast Christian feature stories, preaching and teaching to various fundamentalist and Pentecostal denominations and playing Traditional black gospel, Southern gospel, and inspirational music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_366", "text": "Clooney lists 9 hotels including The Dorchester, 45 Park Lane, Coworth Park, The Beverly Hills Hotel, Hotel Bel-Air, Le Meurice, Hotel Plaza Athenee, Hotel Eden and Hotel Principe di Savoia and asks readers to consider how we are putting money directly into the pockets of men who choose to stone and whip to death their own citizens for being gay or accused of adultery.", "requested_types": ["AWARD", "LOCATION", "MUSICAL_ARTIST", "PERSON"], "distractors": ["AWARD", "MUSICAL_ARTIST"], "gold": {"AWARD": [], "LOCATION": ["The Dorchester", "45 Park Lane", "Coworth Park", "The Beverly Hills Hotel", "Hotel Bel-Air", "Le Meurice", "Hotel Plaza Athenee", "Hotel Eden", "Hotel Principe di Savoia"], "MUSICAL_ARTIST": [], "PERSON": ["Clooney"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, MUSICAL_ARTIST, PERSON\n\nText:\nClooney lists 9 hotels including The Dorchester, 45 Park Lane, Coworth Park, The Beverly Hills Hotel, Hotel Bel-Air, Le Meurice, Hotel Plaza Athenee, Hotel Eden and Hotel Principe di Savoia and asks readers to consider how we are putting money directly into the pockets of men who choose to stone and whip to death their own citizens for being gay or accused of adultery.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, MUSICAL_ARTIST, PERSON\n\nText:\nClooney lists 9 hotels including The Dorchester, 45 Park Lane, Coworth Park, The Beverly Hills Hotel, Hotel Bel-Air, Le Meurice, Hotel Plaza Athenee, Hotel Eden and Hotel Principe di Savoia and asks readers to consider how we are putting money directly into the pockets of men who choose to stone and whip to death their own citizens for being gay or accused of adultery.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_367", "text": "In 2019 Ulvaeus worked with Swedish songwriter Andreas Carlsson to arrange an English dub of Tomas Ledin jukebox musical film En del av mitt hjärta (English: A Piece of My Heart) directed by Edward af Sillén.", "requested_types": ["SONG", "PERSON", "MISC", "MUSICAL_ARTIST", "LOCATION"], "distractors": ["LOCATION"], "gold": {"SONG": ["En del av mitt hjärta", "A Piece of My Heart"], "PERSON": ["Edward af Sillén"], "MISC": ["Swedish", "English", "jukebox musical"], "MUSICAL_ARTIST": ["Ulvaeus", "Andreas Carlsson", "Tomas Ledin"], "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: SONG, PERSON, MISC, MUSICAL_ARTIST, LOCATION\n\nText:\nIn 2019 Ulvaeus worked with Swedish songwriter Andreas Carlsson to arrange an English dub of Tomas Ledin jukebox musical film En del av mitt hjärta (English: A Piece of My Heart) directed by Edward af Sillén.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, PERSON, MISC, MUSICAL_ARTIST, LOCATION\n\nText:\nIn 2019 Ulvaeus worked with Swedish songwriter Andreas Carlsson to arrange an English dub of Tomas Ledin jukebox musical film En del av mitt hjärta (English: A Piece of My Heart) directed by Edward af Sillén.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_368", "text": "Prior to recording, Dylan previewed the songs that would constitute Blood on the Tracks for a number of friends in the music world, including David Crosby, Graham Nash, Stephen Stills, Tim Drummond and Peter Rowan.", "requested_types": ["ALBUM", "MUSICAL_ARTIST", "SONG"], "distractors": ["SONG"], "gold": {"ALBUM": ["Blood on the Tracks"], "MUSICAL_ARTIST": ["Dylan", "David Crosby", "Graham Nash", "Stephen Stills", "Tim Drummond", "Peter Rowan"], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_ARTIST, SONG\n\nText:\nPrior to recording, Dylan previewed the songs that would constitute Blood on the Tracks for a number of friends in the music world, including David Crosby, Graham Nash, Stephen Stills, Tim Drummond and Peter Rowan.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_ARTIST, SONG\n\nText:\nPrior to recording, Dylan previewed the songs that would constitute Blood on the Tracks for a number of friends in the music world, including David Crosby, Graham Nash, Stephen Stills, Tim Drummond and Peter Rowan.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_369", "text": "Their third album, Inhuman Rampage (2006), was certified gold by the Recording Industry Association of America (RIAA) and the British Phonographic Industry (BPI); its lead single, Through the Fire and Flames, is their best-known song, and was featured in several video games, including Guitar Hero III: Legends of Rock.", "requested_types": ["EVENT", "LOCATION", "SONG", "ORGANISATION", "MUSICAL_INSTRUMENT", "MISC", "ALBUM"], "distractors": ["MUSICAL_INSTRUMENT", "EVENT", "LOCATION"], "gold": {"EVENT": [], "LOCATION": [], "SONG": ["Through the Fire and Flames"], "ORGANISATION": ["Recording Industry Association of America", "RIAA", "British Phonographic Industry", "BPI"], "MUSICAL_INSTRUMENT": [], "MISC": ["Guitar Hero III: Legends of Rock"], "ALBUM": ["Inhuman Rampage"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, ORGANISATION, MUSICAL_INSTRUMENT, MISC, ALBUM\n\nText:\nTheir third album, Inhuman Rampage (2006), was certified gold by the Recording Industry Association of America (RIAA) and the British Phonographic Industry (BPI); its lead single, Through the Fire and Flames, is their best-known song, and was featured in several video games, including Guitar Hero III: Legends of Rock.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, ORGANISATION, MUSICAL_INSTRUMENT, MISC, ALBUM\n\nText:\nTheir third album, Inhuman Rampage (2006), was certified gold by the Recording Industry Association of America (RIAA) and the British Phonographic Industry (BPI); its lead single, Through the Fire and Flames, is their best-known song, and was featured in several video games, including Guitar Hero III: Legends of Rock.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_370", "text": "Medeski Martin & amp; Wood, Robert Randolph & amp; the Family Band, Galactic, Widespread Panic, Jam Underground, Diazpora, Soulive, and Karl Denson's Tiny Universe all drew heavily from the funk tradition.", "requested_types": ["SONG", "BAND", "MISC", "MUSIC_GENRE", "LOCATION"], "distractors": ["SONG", "LOCATION", "MISC"], "gold": {"SONG": [], "BAND": ["Medeski Martin & amp; Wood", "Robert Randolph & amp; the Family Band", "Galactic", "Widespread Panic", "Jam Underground", "Diazpora", "Soulive", "Karl Denson's Tiny Universe"], "MISC": [], "MUSIC_GENRE": ["funk"], "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: SONG, BAND, MISC, MUSIC_GENRE, LOCATION\n\nText:\nMedeski Martin & amp; Wood, Robert Randolph & amp; the Family Band, Galactic, Widespread Panic, Jam Underground, Diazpora, Soulive, and Karl Denson's Tiny Universe all drew heavily from the funk tradition.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, MISC, MUSIC_GENRE, LOCATION\n\nText:\nMedeski Martin & amp; Wood, Robert Randolph & amp; the Family Band, Galactic, Widespread Panic, Jam Underground, Diazpora, Soulive, and Karl Denson's Tiny Universe all drew heavily from the funk tradition.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_371", "text": "She represented America in the opening ceremony of 2007 Special Olympics World Summer Games singing the song I Know I Can, and in the opening ceremony of Expo 2010 Shanghai China, singing the song Better City, Better Life with Jon B., both songs which she co-wrote with Quincy Jones.", "requested_types": ["BAND", "MUSIC_GENRE", "AWARD", "SONG", "MUSICAL_ARTIST", "EVENT", "COUNTRY"], "distractors": ["MUSIC_GENRE", "AWARD", "BAND"], "gold": {"BAND": [], "MUSIC_GENRE": [], "AWARD": [], "SONG": ["I Know I Can", "Better City, Better Life"], "MUSICAL_ARTIST": ["Jon B.", "Quincy Jones"], "EVENT": ["2007 Special Olympics World Summer Games", "Expo 2010 Shanghai China"], "COUNTRY": ["America"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, AWARD, SONG, MUSICAL_ARTIST, EVENT, COUNTRY\n\nText:\nShe represented America in the opening ceremony of 2007 Special Olympics World Summer Games singing the song I Know I Can, and in the opening ceremony of Expo 2010 Shanghai China, singing the song Better City, Better Life with Jon B., both songs which she co-wrote with Quincy Jones.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, AWARD, SONG, MUSICAL_ARTIST, EVENT, COUNTRY\n\nText:\nShe represented America in the opening ceremony of 2007 Special Olympics World Summer Games singing the song I Know I Can, and in the opening ceremony of Expo 2010 Shanghai China, singing the song Better City, Better Life with Jon B., both songs which she co-wrote with Quincy Jones.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_372", "text": "Steve Huey of AllMusic said that the band's earlier Punk rock influences are rarely detectable, replaced by surprisingly effective appropriations of Pop music and psychedelia.", "requested_types": ["MISC", "PERSON", "ORGANISATION", "AWARD", "MUSICAL_ARTIST", "MUSIC_GENRE"], "distractors": ["MISC", "PERSON", "AWARD"], "gold": {"MISC": [], "PERSON": [], "ORGANISATION": ["AllMusic"], "AWARD": [], "MUSICAL_ARTIST": ["Steve Huey"], "MUSIC_GENRE": ["Punk rock", "Pop music", "psychedelia"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ORGANISATION, AWARD, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nSteve Huey of AllMusic said that the band's earlier Punk rock influences are rarely detectable, replaced by surprisingly effective appropriations of Pop music and psychedelia.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ORGANISATION, AWARD, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nSteve Huey of AllMusic said that the band's earlier Punk rock influences are rarely detectable, replaced by surprisingly effective appropriations of Pop music and psychedelia.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_373", "text": "Delta blues was also an inspiration for the creation of British skiffle music, from which eventually came the British invasion bands, while simultaneously influencing British blues, which led to the birth of early hard rock and Heavy metal music.", "requested_types": ["MUSICAL_ARTIST", "MISC", "MUSIC_GENRE", "MUSICAL_INSTRUMENT"], "distractors": ["MUSICAL_INSTRUMENT", "MUSICAL_ARTIST"], "gold": {"MUSICAL_ARTIST": [], "MISC": ["British", "British invasion"], "MUSIC_GENRE": ["Delta blues", "skiffle music", "British blues", "hard rock", "Heavy metal music"], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MISC, MUSIC_GENRE, MUSICAL_INSTRUMENT\n\nText:\nDelta blues was also an inspiration for the creation of British skiffle music, from which eventually came the British invasion bands, while simultaneously influencing British blues, which led to the birth of early hard rock and Heavy metal music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MISC, MUSIC_GENRE, MUSICAL_INSTRUMENT\n\nText:\nDelta blues was also an inspiration for the creation of British skiffle music, from which eventually came the British invasion bands, while simultaneously influencing British blues, which led to the birth of early hard rock and Heavy metal music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_374", "text": "Finn has recorded four solo albums, Try Whistling This (1998), One Nil (2001), Dizzy Heights (2014), and Out of Silence (2017).", "requested_types": ["ALBUM", "MUSICAL_ARTIST", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"ALBUM": ["Try Whistling This", "One Nil", "Dizzy Heights", "Out of Silence"], "MUSICAL_ARTIST": ["Finn"], "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: ALBUM, MUSICAL_ARTIST, COUNTRY\n\nText:\nFinn has recorded four solo albums, Try Whistling This (1998), One Nil (2001), Dizzy Heights (2014), and Out of Silence (2017).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_ARTIST, COUNTRY\n\nText:\nFinn has recorded four solo albums, Try Whistling This (1998), One Nil (2001), Dizzy Heights (2014), and Out of Silence (2017).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_375", "text": "They have also won two Grammy Award s for Grammy Award for Best Metal Performance, and in 2019 the band were presented a Grammy Lifetime Achievement Award.", "requested_types": ["BAND", "AWARD"], "distractors": ["BAND"], "gold": {"BAND": [], "AWARD": ["Grammy Award", "Grammy Award for Best Metal Performance", "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: BAND, AWARD\n\nText:\nThey have also won two Grammy Award s for Grammy Award for Best Metal Performance, and in 2019 the band were presented a Grammy Lifetime Achievement 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: BAND, AWARD\n\nText:\nThey have also won two Grammy Award s for Grammy Award for Best Metal Performance, and in 2019 the band were presented a Grammy Lifetime Achievement 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_music_376", "text": "A breakthrough came when Van Damm began to incorporate glamorous nude females on stage, inspired by the Folies Bergère and Moulin Rouge in Paris.", "requested_types": ["PERSON", "ALBUM", "LOCATION"], "distractors": ["ALBUM"], "gold": {"PERSON": ["Van Damm"], "ALBUM": [], "LOCATION": ["Folies Bergère", "Moulin Rouge", "Paris"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, LOCATION\n\nText:\nA breakthrough came when Van Damm began to incorporate glamorous nude females on stage, inspired by the Folies Bergère and Moulin Rouge in Paris.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, LOCATION\n\nText:\nA breakthrough came when Van Damm began to incorporate glamorous nude females on stage, inspired by the Folies Bergère and Moulin Rouge in Paris.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_377", "text": "Since 1993, Jimmy Webb has produced five critically acclaimed solo albums: Suspending Disbelief (1993), Ten Easy Pieces (1996), Twilight of the Renegades (2005), Just Across the River (2010), and Still Within the Sound of My Voice (2013).", "requested_types": ["MISC", "AWARD", "SONG", "MUSICAL_ARTIST", "ALBUM"], "distractors": ["SONG", "AWARD"], "gold": {"MISC": ["solo albums"], "AWARD": [], "SONG": [], "MUSICAL_ARTIST": ["Jimmy Webb"], "ALBUM": ["Suspending Disbelief", "Ten Easy Pieces", "Twilight of the Renegades", "Just Across the River", "Still Within the Sound of My Voice"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST, ALBUM\n\nText:\nSince 1993, Jimmy Webb has produced five critically acclaimed solo albums: Suspending Disbelief (1993), Ten Easy Pieces (1996), Twilight of the Renegades (2005), Just Across the River (2010), and Still Within the Sound of My Voice (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: MISC, AWARD, SONG, MUSICAL_ARTIST, ALBUM\n\nText:\nSince 1993, Jimmy Webb has produced five critically acclaimed solo albums: Suspending Disbelief (1993), Ten Easy Pieces (1996), Twilight of the Renegades (2005), Just Across the River (2010), and Still Within the Sound of My Voice (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_music_378", "text": "By the time he released his debut album 2Pacalypse Now in 1991, he had become a central figure in West Coast hip hop, introducing social issues in the genre at a time when gangsta rap was dominant in the mainstream.", "requested_types": ["PERSON", "ALBUM", "MUSIC_GENRE", "SONG", "BAND"], "distractors": ["BAND", "SONG", "PERSON"], "gold": {"PERSON": [], "ALBUM": ["2Pacalypse Now"], "MUSIC_GENRE": ["West Coast hip hop", "gangsta rap"], "SONG": [], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MUSIC_GENRE, SONG, BAND\n\nText:\nBy the time he released his debut album 2Pacalypse Now in 1991, he had become a central figure in West Coast hip hop, introducing social issues in the genre at a time when gangsta rap was dominant in the mainstream.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, MUSIC_GENRE, SONG, BAND\n\nText:\nBy the time he released his debut album 2Pacalypse Now in 1991, he had become a central figure in West Coast hip hop, introducing social issues in the genre at a time when gangsta rap was dominant in the mainstream.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_379", "text": "Crust punk groups, such as Antisect, Sacrilege and Anti System took some influence from early black metal bands like Venom, Hellhammer, and Celtic Frost, In addition, Norwegian band Darkthrone have incorporated crust punk traits in their more recent material.", "requested_types": ["BAND", "COUNTRY", "SONG", "LOCATION", "MISC", "MUSIC_GENRE"], "distractors": ["COUNTRY", "SONG", "LOCATION"], "gold": {"BAND": ["Antisect", "Sacrilege", "Anti System", "Venom", "Hellhammer", "Celtic Frost", "Darkthrone"], "COUNTRY": [], "SONG": [], "LOCATION": [], "MISC": ["Norwegian"], "MUSIC_GENRE": ["Crust punk", "early black metal", "crust punk"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, COUNTRY, SONG, LOCATION, MISC, MUSIC_GENRE\n\nText:\nCrust punk groups, such as Antisect, Sacrilege and Anti System took some influence from early black metal bands like Venom, Hellhammer, and Celtic Frost, In addition, Norwegian band Darkthrone have incorporated crust punk traits in their more recent material.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, COUNTRY, SONG, LOCATION, MISC, MUSIC_GENRE\n\nText:\nCrust punk groups, such as Antisect, Sacrilege and Anti System took some influence from early black metal bands like Venom, Hellhammer, and Celtic Frost, In addition, Norwegian band Darkthrone have incorporated crust punk traits in their more recent material.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_380", "text": "Brad Shoup of Stereogum surmised that, thanks to the Ramones' praise for the group, many punk, pop punk, or punk-adjacent artists showed influence from the Beach Boys, noting cover versions of the band's songs recorded by Slickee Boys, Agent Orange, Bad Religion, Shonen Knife, the Queers, Hi-Standard, the Descendents, the Donnas, M.O.D., and the Vandals.", "requested_types": ["BAND", "MUSIC_GENRE", "LOCATION", "MISC", "PERSON", "MUSICAL_INSTRUMENT"], "distractors": ["MUSICAL_INSTRUMENT", "LOCATION"], "gold": {"BAND": ["Ramones", "the Beach Boys", "Slickee Boys", "Agent Orange", "Bad Religion", "Shonen Knife", "the Queers", "Hi-Standard", "the Descendents", "the Donnas", "M.O.D.", "the Vandals"], "MUSIC_GENRE": ["punk", "pop punk", "punk-adjacent"], "LOCATION": [], "MISC": ["Stereogum"], "PERSON": ["Brad Shoup"], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, LOCATION, MISC, PERSON, MUSICAL_INSTRUMENT\n\nText:\nBrad Shoup of Stereogum surmised that, thanks to the Ramones' praise for the group, many punk, pop punk, or punk-adjacent artists showed influence from the Beach Boys, noting cover versions of the band's songs recorded by Slickee Boys, Agent Orange, Bad Religion, Shonen Knife, the Queers, Hi-Standard, the Descendents, the Donnas, M.O.D., and the Vandals.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSIC_GENRE, LOCATION, MISC, PERSON, MUSICAL_INSTRUMENT\n\nText:\nBrad Shoup of Stereogum surmised that, thanks to the Ramones' praise for the group, many punk, pop punk, or punk-adjacent artists showed influence from the Beach Boys, noting cover versions of the band's songs recorded by Slickee Boys, Agent Orange, Bad Religion, Shonen Knife, the Queers, Hi-Standard, the Descendents, the Donnas, M.O.D., and the Vandals.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_381", "text": "She also serves on the board of People for the American Way, and volunteers at Amnesty International and Citymeals-on-Wheels.", "requested_types": ["ORGANISATION", "EVENT"], "distractors": ["EVENT"], "gold": {"ORGANISATION": ["People for the American Way", "Amnesty International", "Citymeals-on-Wheels"], "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:\nShe also serves on the board of People for the American Way, and volunteers at Amnesty International and Citymeals-on-Wheels.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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:\nShe also serves on the board of People for the American Way, and volunteers at Amnesty International and Citymeals-on-Wheels.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_382", "text": "Pioneered by black-doom bands like Ophthalamia, Katatonia, Bethlehem, Forgotten Tomb and Shining, depressive suicidal black metal, also known as suicidal black metal, depressive black metal or DSBM, is a style that melds the second wave -style of black metal with doom metal,", "requested_types": ["COUNTRY", "MUSICAL_INSTRUMENT", "MUSIC_GENRE", "BAND"], "distractors": ["COUNTRY", "MUSICAL_INSTRUMENT"], "gold": {"COUNTRY": [], "MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": ["black-doom", "depressive suicidal black metal", "suicidal black metal", "depressive black metal", "DSBM", "black metal", "doom metal"], "BAND": ["Ophthalamia", "Katatonia", "Bethlehem", "Forgotten Tomb", "Shining"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, MUSIC_GENRE, BAND\n\nText:\nPioneered by black-doom bands like Ophthalamia, Katatonia, Bethlehem, Forgotten Tomb and Shining, depressive suicidal black metal, also known as suicidal black metal, depressive black metal or DSBM, is a style that melds the second wave -style of black metal with doom metal,\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, MUSIC_GENRE, BAND\n\nText:\nPioneered by black-doom bands like Ophthalamia, Katatonia, Bethlehem, Forgotten Tomb and Shining, depressive suicidal black metal, also known as suicidal black metal, depressive black metal or DSBM, is a style that melds the second wave -style of black metal with doom metal,\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_383", "text": "A third UK tour for 2017 / 2018 opened at the Curve in Leicester, and also toured to the Birmingham Hippodrome, the Bord Gáis Energy Theatre in Dublin, the Wales Millennium Centre in Cardiff, the Edinburgh Festival Theatre, the Mayflower Theatre in Southampton and the Palace Theatre in Manchester.", "requested_types": ["EVENT", "LOCATION", "ORGANISATION", "COUNTRY"], "distractors": ["EVENT"], "gold": {"EVENT": [], "LOCATION": ["Leicester", "Birmingham Hippodrome", "Bord Gáis Energy Theatre", "Dublin", "Wales Millennium Centre", "Cardiff", "Edinburgh Festival Theatre", "Mayflower Theatre", "Southampton", "Palace Theatre", "Manchester"], "ORGANISATION": ["Curve"], "COUNTRY": ["UK"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY\n\nText:\nA third UK tour for 2017 / 2018 opened at the Curve in Leicester, and also toured to the Birmingham Hippodrome, the Bord Gáis Energy Theatre in Dublin, the Wales Millennium Centre in Cardiff, the Edinburgh Festival Theatre, the Mayflower Theatre in Southampton and the Palace Theatre in Manchester.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY\n\nText:\nA third UK tour for 2017 / 2018 opened at the Curve in Leicester, and also toured to the Birmingham Hippodrome, the Bord Gáis Energy Theatre in Dublin, the Wales Millennium Centre in Cardiff, the Edinburgh Festival Theatre, the Mayflower Theatre in Southampton and the Palace Theatre in Manchester.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_384", "text": "Arthur Schwartz and Howard Dietz first tried, and then Richard Rodgers and Oscar Hammerstein II attempted, but gave up and Hammerstein told Lerner, Pygmalion had no subplot.", "requested_types": ["COUNTRY", "MISC", "MUSICAL_ARTIST"], "distractors": ["COUNTRY"], "gold": {"COUNTRY": [], "MISC": ["Pygmalion"], "MUSICAL_ARTIST": ["Arthur Schwartz", "Howard Dietz", "Richard Rodgers", "Oscar Hammerstein II", "Hammerstein", "Lerner"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST\n\nText:\nArthur Schwartz and Howard Dietz first tried, and then Richard Rodgers and Oscar Hammerstein II attempted, but gave up and Hammerstein told Lerner, Pygmalion had no subplot.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST\n\nText:\nArthur Schwartz and Howard Dietz first tried, and then Richard Rodgers and Oscar Hammerstein II attempted, but gave up and Hammerstein told Lerner, Pygmalion had no subplot.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_385", "text": "Van Bentum competed in three consecutive Summer Olympics for her native country, starting in Swimming at the 1980 Summer Olympics.", "requested_types": ["PERSON", "MUSIC_GENRE", "EVENT", "BAND"], "distractors": ["MUSIC_GENRE", "BAND"], "gold": {"PERSON": ["Van Bentum"], "MUSIC_GENRE": [], "EVENT": ["Summer Olympics", "1980 Summer Olympics"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, EVENT, BAND\n\nText:\nVan Bentum competed in three consecutive Summer Olympics for her native country, starting in Swimming at the 1980 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: PERSON, MUSIC_GENRE, EVENT, BAND\n\nText:\nVan Bentum competed in three consecutive Summer Olympics for her native country, starting in Swimming at the 1980 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_music_386", "text": "Morissette assumed creative control and producing duties for her subsequent studio albums, including Under Rug Swept (2002), So-Called Chaos (2004), and Flavors of Entanglement (2008).", "requested_types": ["MISC", "MUSICAL_ARTIST", "ALBUM"], "distractors": ["MISC"], "gold": {"MISC": [], "MUSICAL_ARTIST": ["Morissette"], "ALBUM": ["Under Rug Swept", "So-Called Chaos", "Flavors of Entanglement"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ALBUM\n\nText:\nMorissette assumed creative control and producing duties for her subsequent studio albums, including Under Rug Swept (2002), So-Called Chaos (2004), and Flavors of Entanglement (2008).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ALBUM\n\nText:\nMorissette assumed creative control and producing duties for her subsequent studio albums, including Under Rug Swept (2002), So-Called Chaos (2004), and Flavors of Entanglement (2008).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_387", "text": "He also recorded jazz, such as The Great Ray Charles (1957) and worked with vibraphonist Milt Jackson, releasing Soul Brothers in 1958 and Soul Meeting in 1961.", "requested_types": ["MUSIC_GENRE", "MUSICAL_ARTIST", "ALBUM", "BAND"], "distractors": ["BAND"], "gold": {"MUSIC_GENRE": ["jazz"], "MUSICAL_ARTIST": ["Milt Jackson"], "ALBUM": ["The Great Ray Charles", "Soul Brothers", "Soul Meeting"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, ALBUM, BAND\n\nText:\nHe also recorded jazz, such as The Great Ray Charles (1957) and worked with vibraphonist Milt Jackson, releasing Soul Brothers in 1958 and Soul Meeting in 1961.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, ALBUM, BAND\n\nText:\nHe also recorded jazz, such as The Great Ray Charles (1957) and worked with vibraphonist Milt Jackson, releasing Soul Brothers in 1958 and Soul Meeting in 1961.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_388", "text": "The groups which made up the Native Tongues Posse tended toward jazzy releases: these include the Jungle Brothers' debut Straight Out the Jungle (1988), and A Tribe Called Quest 's People's Instinctive Travels and the Paths of Rhythm (1990) and The Low End Theory (1991).", "requested_types": ["ALBUM", "SONG", "MISC", "BAND"], "distractors": ["SONG"], "gold": {"ALBUM": ["Straight Out the Jungle", "People's Instinctive Travels and the Paths of Rhythm", "The Low End Theory"], "SONG": [], "MISC": ["Native Tongues Posse", "jazzy"], "BAND": ["Jungle Brothers", "A Tribe Called Quest"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, SONG, MISC, BAND\n\nText:\nThe groups which made up the Native Tongues Posse tended toward jazzy releases: these include the Jungle Brothers' debut Straight Out the Jungle (1988), and A Tribe Called Quest 's People's Instinctive Travels and the Paths of Rhythm (1990) and The Low End Theory (1991).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, SONG, MISC, BAND\n\nText:\nThe groups which made up the Native Tongues Posse tended toward jazzy releases: these include the Jungle Brothers' debut Straight Out the Jungle (1988), and A Tribe Called Quest 's People's Instinctive Travels and the Paths of Rhythm (1990) and The Low End Theory (1991).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_389", "text": "The music arose as a synthesis of traditional Creole music, some Cajun music influences, and African-American traditions, including Rhythm and blues, blues, jazz, and Gospel music.", "requested_types": ["ORGANISATION", "MUSIC_GENRE", "PERSON", "COUNTRY", "MISC"], "distractors": ["PERSON", "ORGANISATION", "COUNTRY"], "gold": {"ORGANISATION": [], "MUSIC_GENRE": ["Creole music", "Cajun music", "Rhythm and blues", "blues", "jazz", "Gospel music"], "PERSON": [], "COUNTRY": [], "MISC": ["African-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: ORGANISATION, MUSIC_GENRE, PERSON, COUNTRY, MISC\n\nText:\nThe music arose as a synthesis of traditional Creole music, some Cajun music influences, and African-American traditions, including Rhythm and blues, blues, jazz, and Gospel music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, PERSON, COUNTRY, MISC\n\nText:\nThe music arose as a synthesis of traditional Creole music, some Cajun music influences, and African-American traditions, including Rhythm and blues, blues, jazz, and Gospel music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_390", "text": "Lyngstad returned to the recording studio in 2010 to record vocals for the Cat Stevens song Morning Has Broken, for Swedish guitarist Georg Wadenius' s album Reconnection.", "requested_types": ["MISC", "SONG", "MUSIC_GENRE", "ALBUM", "BAND", "MUSICAL_ARTIST", "EVENT"], "distractors": ["MUSIC_GENRE", "BAND", "EVENT"], "gold": {"MISC": ["Swedish"], "SONG": ["Morning Has Broken"], "MUSIC_GENRE": [], "ALBUM": ["Reconnection"], "BAND": [], "MUSICAL_ARTIST": ["Lyngstad", "Cat Stevens", "Georg Wadenius"], "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, SONG, MUSIC_GENRE, ALBUM, BAND, MUSICAL_ARTIST, EVENT\n\nText:\nLyngstad returned to the recording studio in 2010 to record vocals for the Cat Stevens song Morning Has Broken, for Swedish guitarist Georg Wadenius' s album Reconnection.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE, ALBUM, BAND, MUSICAL_ARTIST, EVENT\n\nText:\nLyngstad returned to the recording studio in 2010 to record vocals for the Cat Stevens song Morning Has Broken, for Swedish guitarist Georg Wadenius' s album Reconnection.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_391", "text": "The album was a commercial disappointment compared to the band's previous effort, though it charted generally high at # 5 in the United Kingdom and Germany, # 7 on the Austria n and Switzerland music charts and # 8 in Norway.", "requested_types": ["AWARD", "MUSICAL_ARTIST", "MISC", "COUNTRY"], "distractors": ["MUSICAL_ARTIST", "AWARD", "MISC"], "gold": {"AWARD": [], "MUSICAL_ARTIST": [], "MISC": [], "COUNTRY": ["United Kingdom", "Germany", "Austria", "Switzerland", "Norway"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, MISC, COUNTRY\n\nText:\nThe album was a commercial disappointment compared to the band's previous effort, though it charted generally high at # 5 in the United Kingdom and Germany, # 7 on the Austria n and Switzerland music charts and # 8 in Norway.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, MISC, COUNTRY\n\nText:\nThe album was a commercial disappointment compared to the band's previous effort, though it charted generally high at # 5 in the United Kingdom and Germany, # 7 on the Austria n and Switzerland music charts and # 8 in Norway.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_392", "text": "Returning to his solo work, Iommi enlisted bassist Dave Spitz (ex- Great White), drummer Eric Singer and initially intended to use multiple singers, including Rob Halford of Judas Priest, former Deep Purple and Trapeze vocalist Glenn Hughes, and former Sabbath vocalist Ronnie James Dio.", "requested_types": ["BAND", "MUSICAL_ARTIST", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"BAND": ["Great White", "Judas Priest", "Deep Purple", "Trapeze", "Sabbath"], "MUSICAL_ARTIST": ["Iommi", "Dave Spitz", "Eric Singer", "Rob Halford", "Glenn Hughes", "Ronnie James Dio"], "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: BAND, MUSICAL_ARTIST, COUNTRY\n\nText:\nReturning to his solo work, Iommi enlisted bassist Dave Spitz (ex- Great White), drummer Eric Singer and initially intended to use multiple singers, including Rob Halford of Judas Priest, former Deep Purple and Trapeze vocalist Glenn Hughes, and former Sabbath vocalist Ronnie James Dio.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSICAL_ARTIST, COUNTRY\n\nText:\nReturning to his solo work, Iommi enlisted bassist Dave Spitz (ex- Great White), drummer Eric Singer and initially intended to use multiple singers, including Rob Halford of Judas Priest, former Deep Purple and Trapeze vocalist Glenn Hughes, and former Sabbath vocalist Ronnie James Dio.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_393", "text": "This scene and the associated dances and fashions quickly spread to other UK dancehalls and nightclubs like the Chateau Impney (Droitwich), Catacombs (Wolverhampton), the Highland Rooms at Blackpool Mecca, Golden Torch (Stoke-on-Trent) and Wigan Casino.", "requested_types": ["MUSIC_GENRE", "COUNTRY", "ALBUM", "LOCATION"], "distractors": ["MUSIC_GENRE", "ALBUM"], "gold": {"MUSIC_GENRE": [], "COUNTRY": ["UK"], "ALBUM": [], "LOCATION": ["Chateau Impney", "Droitwich", "Catacombs", "Wolverhampton", "Highland Rooms", "Blackpool Mecca", "Golden Torch", "Stoke-on-Trent", "Wigan Casino"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, COUNTRY, ALBUM, LOCATION\n\nText:\nThis scene and the associated dances and fashions quickly spread to other UK dancehalls and nightclubs like the Chateau Impney (Droitwich), Catacombs (Wolverhampton), the Highland Rooms at Blackpool Mecca, Golden Torch (Stoke-on-Trent) and Wigan Casino.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, COUNTRY, ALBUM, LOCATION\n\nText:\nThis scene and the associated dances and fashions quickly spread to other UK dancehalls and nightclubs like the Chateau Impney (Droitwich), Catacombs (Wolverhampton), the Highland Rooms at Blackpool Mecca, Golden Torch (Stoke-on-Trent) and Wigan Casino.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_394", "text": "The exceptions to this scheme were the band's fourth album, a live boxed set entitled Chicago at Carnegie Hall, their twelfth album Hot Streets, and the Arabic-numbered Chicago 13.", "requested_types": ["ALBUM", "COUNTRY"], "distractors": ["COUNTRY"], "gold": {"ALBUM": ["Chicago at Carnegie Hall", "Hot Streets", "Chicago 13"], "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: ALBUM, COUNTRY\n\nText:\nThe exceptions to this scheme were the band's fourth album, a live boxed set entitled Chicago at Carnegie Hall, their twelfth album Hot Streets, and the Arabic-numbered Chicago 13.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, COUNTRY\n\nText:\nThe exceptions to this scheme were the band's fourth album, a live boxed set entitled Chicago at Carnegie Hall, their twelfth album Hot Streets, and the Arabic-numbered Chicago 13.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_395", "text": "Deacon Blue also performed at the Glasgow 2014 Commonwealth Games closing ceremony on 3 August 2014, performing their hit, Dignity.", "requested_types": ["PERSON", "MUSICAL_INSTRUMENT", "EVENT", "LOCATION", "BAND", "SONG"], "distractors": ["PERSON", "MUSICAL_INSTRUMENT"], "gold": {"PERSON": [], "MUSICAL_INSTRUMENT": [], "EVENT": ["2014 Commonwealth Games"], "LOCATION": ["Glasgow"], "BAND": ["Deacon Blue"], "SONG": ["Dignity"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, EVENT, LOCATION, BAND, SONG\n\nText:\nDeacon Blue also performed at the Glasgow 2014 Commonwealth Games closing ceremony on 3 August 2014, performing their hit, Dignity.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, EVENT, LOCATION, BAND, SONG\n\nText:\nDeacon Blue also performed at the Glasgow 2014 Commonwealth Games closing ceremony on 3 August 2014, performing their hit, Dignity.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_396", "text": "The band released seven studio albums between 1994 and 2005: Sleater-Kinney (1995), Call the Doctor (1996), Dig Me Out (1997), The Hot Rock (1999), All Hands on the Bad One (2000), One Beat (2002) and The Woods (2005).", "requested_types": ["BAND", "ALBUM", "COUNTRY"], "distractors": ["BAND", "COUNTRY"], "gold": {"BAND": [], "ALBUM": ["Sleater-Kinney", "Call the Doctor", "Dig Me Out", "The Hot Rock", "All Hands on the Bad One", "One Beat", "The Woods"], "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: BAND, ALBUM, COUNTRY\n\nText:\nThe band released seven studio albums between 1994 and 2005: Sleater-Kinney (1995), Call the Doctor (1996), Dig Me Out (1997), The Hot Rock (1999), All Hands on the Bad One (2000), One Beat (2002) and The Woods (2005).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, ALBUM, COUNTRY\n\nText:\nThe band released seven studio albums between 1994 and 2005: Sleater-Kinney (1995), Call the Doctor (1996), Dig Me Out (1997), The Hot Rock (1999), All Hands on the Bad One (2000), One Beat (2002) and The Woods (2005).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_397", "text": "The film was nominated for Academy Awards for Academy Award for Best Costume Design, Academy Award for Best Original Score and Academy Award for Best Original Screenplay.", "requested_types": ["AWARD", "SONG", "MUSICAL_INSTRUMENT"], "distractors": ["SONG", "MUSICAL_INSTRUMENT"], "gold": {"AWARD": ["Academy Awards", "Academy Award for Best Costume Design", "Academy Award for Best Original Score", "Academy Award for Best Original Screenplay"], "SONG": [], "MUSICAL_INSTRUMENT": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_INSTRUMENT\n\nText:\nThe film was nominated for Academy Awards for Academy Award for Best Costume Design, Academy Award for Best Original Score and Academy Award for Best Original Screenplay.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_INSTRUMENT\n\nText:\nThe film was nominated for Academy Awards for Academy Award for Best Costume Design, Academy Award for Best Original Score and Academy Award for Best Original Screenplay.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_398", "text": "He hosted the television show Gastank, and recorded his first of several New-age music, ambient, and Christian music albums with Country Airs (1986) and The Gospels (1987), respectively.", "requested_types": ["ALBUM", "LOCATION", "MUSIC_GENRE", "AWARD", "MISC"], "distractors": ["AWARD", "LOCATION"], "gold": {"ALBUM": ["Country Airs", "The Gospels"], "LOCATION": [], "MUSIC_GENRE": ["New-age music", "ambient", "Christian music"], "AWARD": [], "MISC": ["Gastank"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, LOCATION, MUSIC_GENRE, AWARD, MISC\n\nText:\nHe hosted the television show Gastank, and recorded his first of several New-age music, ambient, and Christian music albums with Country Airs (1986) and The Gospels (1987), 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: ALBUM, LOCATION, MUSIC_GENRE, AWARD, MISC\n\nText:\nHe hosted the television show Gastank, and recorded his first of several New-age music, ambient, and Christian music albums with Country Airs (1986) and The Gospels (1987), 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_music_399", "text": "Heavily based in soul music, neo soul is distinguished by a less conventional sound than its contemporary R & B counterpart, with incorporated elements ranging from jazz, funk, Hip hop music and Electronic music to Pop music, Jazz fusion, and African music.", "requested_types": ["MUSIC_GENRE", "COUNTRY", "ORGANISATION"], "distractors": ["ORGANISATION", "COUNTRY"], "gold": {"MUSIC_GENRE": ["soul music", "neo soul", "R & B", "jazz", "funk", "Hip hop music", "Electronic music", "Pop music", "Jazz fusion", "African music"], "COUNTRY": [], "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: MUSIC_GENRE, COUNTRY, ORGANISATION\n\nText:\nHeavily based in soul music, neo soul is distinguished by a less conventional sound than its contemporary R & B counterpart, with incorporated elements ranging from jazz, funk, Hip hop music and Electronic music to Pop music, Jazz fusion, and African music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, COUNTRY, ORGANISATION\n\nText:\nHeavily based in soul music, neo soul is distinguished by a less conventional sound than its contemporary R & B counterpart, with incorporated elements ranging from jazz, funk, Hip hop music and Electronic music to Pop music, Jazz fusion, and African music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_400", "text": "The Locust, from San Diego, In Los Angeles, Hole also initially drew influence from grindcore in their early releases, particularly on their singles Dicknail and Teenage Whore, as well as on their debut album, Pretty on the Inside (1991), {{cite AV media notes", "requested_types": ["SONG", "MUSIC_GENRE", "MUSICAL_ARTIST", "ALBUM", "BAND", "LOCATION"], "distractors": ["MUSICAL_ARTIST"], "gold": {"SONG": ["Dicknail", "Teenage Whore"], "MUSIC_GENRE": ["grindcore"], "MUSICAL_ARTIST": [], "ALBUM": ["Pretty on the Inside"], "BAND": ["The Locust", "Hole"], "LOCATION": ["San Diego", "Los Angeles"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE, MUSICAL_ARTIST, ALBUM, BAND, LOCATION\n\nText:\nThe Locust, from San Diego, In Los Angeles, Hole also initially drew influence from grindcore in their early releases, particularly on their singles Dicknail and Teenage Whore, as well as on their debut album, Pretty on the Inside (1991), {{cite AV media notes\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE, MUSICAL_ARTIST, ALBUM, BAND, LOCATION\n\nText:\nThe Locust, from San Diego, In Los Angeles, Hole also initially drew influence from grindcore in their early releases, particularly on their singles Dicknail and Teenage Whore, as well as on their debut album, Pretty on the Inside (1991), {{cite AV media notes\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_401", "text": "In English skiffle bands, Australia n and New Zealand bush band s and South Africa n kwela bands, the same sort of bass has a tea chest as a resonator.", "requested_types": ["COUNTRY", "MUSIC_GENRE", "BAND", "PERSON", "MISC"], "distractors": ["PERSON"], "gold": {"COUNTRY": ["Australia", "New Zealand", "South Africa"], "MUSIC_GENRE": ["skiffle", "kwela"], "BAND": ["bush"], "PERSON": [], "MISC": ["English", "tea chest"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, BAND, PERSON, MISC\n\nText:\nIn English skiffle bands, Australia n and New Zealand bush band s and South Africa n kwela bands, the same sort of bass has a tea chest as a resonator.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, BAND, PERSON, MISC\n\nText:\nIn English skiffle bands, Australia n and New Zealand bush band s and South Africa n kwela bands, the same sort of bass has a tea chest as a resonator.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_402", "text": "For example, in Major Minus and Hurts Like Heaven, both from the album Mylo Xyloto (2011), Chris Martin 's vocals are mostly vocoder-processed.", "requested_types": ["SONG", "COUNTRY", "ALBUM", "LOCATION", "MUSICAL_ARTIST"], "distractors": ["LOCATION", "COUNTRY"], "gold": {"SONG": ["Major Minus", "Hurts Like Heaven"], "COUNTRY": [], "ALBUM": ["Mylo Xyloto"], "LOCATION": [], "MUSICAL_ARTIST": ["Chris Martin"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, COUNTRY, ALBUM, LOCATION, MUSICAL_ARTIST\n\nText:\nFor example, in Major Minus and Hurts Like Heaven, both from the album Mylo Xyloto (2011), Chris Martin 's vocals are mostly vocoder-processed.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, COUNTRY, ALBUM, LOCATION, MUSICAL_ARTIST\n\nText:\nFor example, in Major Minus and Hurts Like Heaven, both from the album Mylo Xyloto (2011), Chris Martin 's vocals are mostly vocoder-processed.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_403", "text": "Major artists of the Texas style are Johnny Winter, Stevie Ray Vaughan, the The Fabulous Thunderbirds (led by harmonica player and singer-songwriter Kim Wilson), and ZZ Top.", "requested_types": ["MISC", "BAND", "ALBUM", "COUNTRY", "MUSIC_GENRE", "MUSICAL_INSTRUMENT", "MUSICAL_ARTIST"], "distractors": ["MUSIC_GENRE", "COUNTRY", "ALBUM"], "gold": {"MISC": ["Texas style"], "BAND": ["The Fabulous Thunderbirds", "ZZ Top"], "ALBUM": [], "COUNTRY": [], "MUSIC_GENRE": [], "MUSICAL_INSTRUMENT": ["harmonica"], "MUSICAL_ARTIST": ["Johnny Winter", "Stevie Ray Vaughan", "Kim Wilson"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM, COUNTRY, MUSIC_GENRE, MUSICAL_INSTRUMENT, MUSICAL_ARTIST\n\nText:\nMajor artists of the Texas style are Johnny Winter, Stevie Ray Vaughan, the The Fabulous Thunderbirds (led by harmonica player and singer-songwriter Kim Wilson), and ZZ Top.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM, COUNTRY, MUSIC_GENRE, MUSICAL_INSTRUMENT, MUSICAL_ARTIST\n\nText:\nMajor artists of the Texas style are Johnny Winter, Stevie Ray Vaughan, the The Fabulous Thunderbirds (led by harmonica player and singer-songwriter Kim Wilson), and ZZ Top.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_404", "text": "A recipient of many awards and honors, she is one of the few entertainers to have won an Emmy Award, a Grammy Award, an Academy Awards, and a Tony Award.", "requested_types": ["PERSON", "AWARD", "MISC", "ORGANISATION"], "distractors": ["ORGANISATION", "PERSON", "MISC"], "gold": {"PERSON": [], "AWARD": ["Emmy Award", "Grammy Award", "Academy Awards", "Tony Award"], "MISC": [], "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: PERSON, AWARD, MISC, ORGANISATION\n\nText:\nA recipient of many awards and honors, she is one of the few entertainers to have won an Emmy Award, a Grammy Award, an Academy Awards, and a Tony 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: PERSON, AWARD, MISC, ORGANISATION\n\nText:\nA recipient of many awards and honors, she is one of the few entertainers to have won an Emmy Award, a Grammy Award, an Academy Awards, and a Tony 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_music_405", "text": "He worked with Berlin State Opera; La Scala, Milan; Royal Opera Stockholm; the Royal Opera House at Covent Garden, Chorégies d'Orange and Houston Grand Opera.", "requested_types": ["ORGANISATION", "AWARD", "ALBUM", "PERSON", "LOCATION"], "distractors": ["ALBUM", "AWARD", "PERSON"], "gold": {"ORGANISATION": ["Berlin State Opera"], "AWARD": [], "ALBUM": [], "PERSON": [], "LOCATION": ["La Scala", "Milan", "Royal Opera Stockholm", "Royal Opera House", "Covent Garden", "Chorégies d'Orange", "Houston Grand Opera"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, PERSON, LOCATION\n\nText:\nHe worked with Berlin State Opera; La Scala, Milan; Royal Opera Stockholm; the Royal Opera House at Covent Garden, Chorégies d'Orange and Houston Grand Opera.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM, PERSON, LOCATION\n\nText:\nHe worked with Berlin State Opera; La Scala, Milan; Royal Opera Stockholm; the Royal Opera House at Covent Garden, Chorégies d'Orange and Houston Grand Opera.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_406", "text": "Notable recital engagements have included her Carnegie Hall debut and performances at the Kennedy Center, Symphony Center, Symphony Hall, Barbican Centre, Philharmonie, and Concertgebouw.", "requested_types": ["SONG", "LOCATION", "MISC"], "distractors": ["MISC", "SONG"], "gold": {"SONG": [], "LOCATION": ["Carnegie Hall", "Kennedy Center", "Symphony Center", "Symphony Hall", "Barbican Centre", "Philharmonie", "Concertgebouw"], "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: SONG, LOCATION, MISC\n\nText:\nNotable recital engagements have included her Carnegie Hall debut and performances at the Kennedy Center, Symphony Center, Symphony Hall, Barbican Centre, Philharmonie, and Concertgebouw.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, LOCATION, MISC\n\nText:\nNotable recital engagements have included her Carnegie Hall debut and performances at the Kennedy Center, Symphony Center, Symphony Hall, Barbican Centre, Philharmonie, and Concertgebouw.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_407", "text": "He is the recipient of numerous accolades, including a Cannes Film Festival Award for Best Actor and nominations for a Tony Award, an Academy Awards, two Primetime Emmy Award s and five Golden Globe Awards.", "requested_types": ["MISC", "ORGANISATION", "AWARD"], "distractors": ["ORGANISATION", "MISC"], "gold": {"MISC": [], "ORGANISATION": [], "AWARD": ["Cannes Film Festival Award for Best Actor", "Tony Award", "Academy Awards", "Primetime Emmy Award", "Golden Globe Awards"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, AWARD\n\nText:\nHe is the recipient of numerous accolades, including a Cannes Film Festival Award for Best Actor and nominations for a Tony Award, an Academy Awards, two Primetime Emmy Award s and five Golden Globe 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: MISC, ORGANISATION, AWARD\n\nText:\nHe is the recipient of numerous accolades, including a Cannes Film Festival Award for Best Actor and nominations for a Tony Award, an Academy Awards, two Primetime Emmy Award s and five Golden Globe 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_music_408", "text": "In this era, Chicago radio jocks The Hot Mix 5, and club DJs Ron Hardy and Frankie Knuckles played various styles of dance music, including older disco records, newer Italo disco and electro funk tracks, B-boy hip hop music by Man Parrish, Jellybean Benitez, Arthur Baker and John Robie as well as Electropop music by Kraftwerk and Yellow Magic Orchestra.", "requested_types": ["LOCATION", "MUSIC_GENRE", "EVENT", "BAND", "MUSICAL_ARTIST", "PERSON"], "distractors": ["EVENT", "PERSON"], "gold": {"LOCATION": ["Chicago"], "MUSIC_GENRE": ["dance music", "disco", "Italo disco", "electro funk", "hip hop music", "Electropop", "music"], "EVENT": [], "BAND": ["Hot Mix 5", "Kraftwerk", "Yellow Magic Orchestra"], "MUSICAL_ARTIST": ["Ron Hardy", "Frankie Knuckles", "Man Parrish", "Jellybean Benitez", "Arthur Baker", "John Robie"], "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, MUSIC_GENRE, EVENT, BAND, MUSICAL_ARTIST, PERSON\n\nText:\nIn this era, Chicago radio jocks The Hot Mix 5, and club DJs Ron Hardy and Frankie Knuckles played various styles of dance music, including older disco records, newer Italo disco and electro funk tracks, B-boy hip hop music by Man Parrish, Jellybean Benitez, Arthur Baker and John Robie as well as Electropop music by Kraftwerk and Yellow Magic Orchestra.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, EVENT, BAND, MUSICAL_ARTIST, PERSON\n\nText:\nIn this era, Chicago radio jocks The Hot Mix 5, and club DJs Ron Hardy and Frankie Knuckles played various styles of dance music, including older disco records, newer Italo disco and electro funk tracks, B-boy hip hop music by Man Parrish, Jellybean Benitez, Arthur Baker and John Robie as well as Electropop music by Kraftwerk and Yellow Magic Orchestra.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_409", "text": "Nine Inch Nails covered the song Metal on The Fragile remix album Things Falling Apart as did Afrika Bambaataa (with Numan himself) on the album Dark Matter Moving at the Speed of Light.", "requested_types": ["SONG", "MUSIC_GENRE", "MUSICAL_ARTIST", "LOCATION", "BAND", "ALBUM", "COUNTRY"], "distractors": ["LOCATION", "COUNTRY", "MUSIC_GENRE"], "gold": {"SONG": ["Metal"], "MUSIC_GENRE": [], "MUSICAL_ARTIST": ["Afrika Bambaataa", "Numan"], "LOCATION": [], "BAND": ["Nine Inch Nails"], "ALBUM": ["The Fragile", "Things Falling Apart", "Dark Matter Moving at the Speed of Light"], "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: SONG, MUSIC_GENRE, MUSICAL_ARTIST, LOCATION, BAND, ALBUM, COUNTRY\n\nText:\nNine Inch Nails covered the song Metal on The Fragile remix album Things Falling Apart as did Afrika Bambaataa (with Numan himself) on the album Dark Matter Moving at the Speed of Light.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSIC_GENRE, MUSICAL_ARTIST, LOCATION, BAND, ALBUM, COUNTRY\n\nText:\nNine Inch Nails covered the song Metal on The Fragile remix album Things Falling Apart as did Afrika Bambaataa (with Numan himself) on the album Dark Matter Moving at the Speed of Light.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_410", "text": "Goodness toured extensively all over the world, supporting such acts as Pearl Jam, Cheap Trick, and Oasis.", "requested_types": ["SONG", "BAND", "MISC", "MUSIC_GENRE"], "distractors": ["MISC", "MUSIC_GENRE", "SONG"], "gold": {"SONG": [], "BAND": ["Goodness", "Pearl Jam", "Cheap Trick", "Oasis"], "MISC": [], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, MISC, MUSIC_GENRE\n\nText:\nGoodness toured extensively all over the world, supporting such acts as Pearl Jam, Cheap Trick, and Oasis.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, MISC, MUSIC_GENRE\n\nText:\nGoodness toured extensively all over the world, supporting such acts as Pearl Jam, Cheap Trick, and Oasis.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_411", "text": "He appeared on several Jack Bruce albums, including A Question of Time, Cities of the Heart, Monkjack and More Jack than God.", "requested_types": ["PERSON", "BAND", "ALBUM", "AWARD", "SONG"], "distractors": ["AWARD", "BAND"], "gold": {"PERSON": ["Jack Bruce"], "BAND": [], "ALBUM": ["Cities of the Heart", "Monkjack", "More Jack than God"], "AWARD": [], "SONG": ["A Question of Time"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM, AWARD, SONG\n\nText:\nHe appeared on several Jack Bruce albums, including A Question of Time, Cities of the Heart, Monkjack and More Jack than God.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, ALBUM, AWARD, SONG\n\nText:\nHe appeared on several Jack Bruce albums, including A Question of Time, Cities of the Heart, Monkjack and More Jack than God.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_412", "text": "Sometimes it is taken to include Greenland and historically the Baltic countries of Estonia, Latvia and Lithuania.", "requested_types": ["SONG", "MISC", "MUSICAL_ARTIST", "COUNTRY", "LOCATION"], "distractors": ["SONG", "MUSICAL_ARTIST", "LOCATION"], "gold": {"SONG": [], "MISC": ["Baltic countries"], "MUSICAL_ARTIST": [], "COUNTRY": ["Greenland", "Estonia", "Latvia", "Lithuania"], "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: SONG, MISC, MUSICAL_ARTIST, COUNTRY, LOCATION\n\nText:\nSometimes it is taken to include Greenland and historically the Baltic countries of Estonia, Latvia and Lithuania.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MISC, MUSICAL_ARTIST, COUNTRY, LOCATION\n\nText:\nSometimes it is taken to include Greenland and historically the Baltic countries of Estonia, Latvia and Lithuania.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_413", "text": "He lent active support to several philanthropic endeavours such as the library at St. Mary's School, Mumbai, Bombay Hospital, Child Rights and You, Save the Children and ALMA.", "requested_types": ["LOCATION", "BAND", "PERSON", "ORGANISATION"], "distractors": ["BAND", "PERSON"], "gold": {"LOCATION": ["St. Mary's School", "Mumbai", "Bombay Hospital"], "BAND": [], "PERSON": [], "ORGANISATION": ["Child Rights and You", "Save the Children", "ALMA"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, PERSON, ORGANISATION\n\nText:\nHe lent active support to several philanthropic endeavours such as the library at St. Mary's School, Mumbai, Bombay Hospital, Child Rights and You, Save the Children and ALMA.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, PERSON, ORGANISATION\n\nText:\nHe lent active support to several philanthropic endeavours such as the library at St. Mary's School, Mumbai, Bombay Hospital, Child Rights and You, Save the Children and ALMA.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_414", "text": "She was a fellow, examiner, lecturer, committee and council member of the Imperial Society of Teachers of Dancing, life member of the Royal Academy of Dance, and Honorary MA.", "requested_types": ["AWARD", "ORGANISATION", "COUNTRY", "BAND"], "distractors": ["COUNTRY", "BAND"], "gold": {"AWARD": ["Honorary MA"], "ORGANISATION": ["Imperial Society of Teachers of Dancing", "Royal Academy of Dance"], "COUNTRY": [], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, BAND\n\nText:\nShe was a fellow, examiner, lecturer, committee and council member of the Imperial Society of Teachers of Dancing, life member of the Royal Academy of Dance, and Honorary MA.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, COUNTRY, BAND\n\nText:\nShe was a fellow, examiner, lecturer, committee and council member of the Imperial Society of Teachers of Dancing, life member of the Royal Academy of Dance, and Honorary MA.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_415", "text": "In addition to Parliament Funkadelic, artists like Sly and the Family Stone, Rufus & Chaka Khan, Bootsy's Rubber Band, the Isley Brothers, Ohio Players, Con Funk Shun, Kool and the Gang, the Bar-Kays, Commodores, Roy Ayers, and Stevie Wonder, among others, were successful in getting radio play.", "requested_types": ["SONG", "BAND", "MUSICAL_ARTIST"], "distractors": ["SONG"], "gold": {"SONG": [], "BAND": ["Parliament Funkadelic", "Sly and the Family Stone", "Rufus & Chaka Khan", "Bootsy's Rubber Band", "Isley Brothers", "Ohio Players", "Con Funk Shun", "Kool and the Gang", "the Bar-Kays", "Commodores"], "MUSICAL_ARTIST": ["Roy Ayers", "Stevie Wonder"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, MUSICAL_ARTIST\n\nText:\nIn addition to Parliament Funkadelic, artists like Sly and the Family Stone, Rufus & Chaka Khan, Bootsy's Rubber Band, the Isley Brothers, Ohio Players, Con Funk Shun, Kool and the Gang, the Bar-Kays, Commodores, Roy Ayers, and Stevie Wonder, among others, were successful in getting radio play.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, MUSICAL_ARTIST\n\nText:\nIn addition to Parliament Funkadelic, artists like Sly and the Family Stone, Rufus & Chaka Khan, Bootsy's Rubber Band, the Isley Brothers, Ohio Players, Con Funk Shun, Kool and the Gang, the Bar-Kays, Commodores, Roy Ayers, and Stevie Wonder, among others, were successful in getting radio play.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_416", "text": "Electropop is a music genre combining elements of Electronic music and Pop music genres.", "requested_types": ["MUSIC_GENRE", "MUSICAL_ARTIST", "SONG", "ALBUM"], "distractors": ["ALBUM", "MUSICAL_ARTIST", "SONG"], "gold": {"MUSIC_GENRE": ["Electropop", "Electronic music", "Pop music"], "MUSICAL_ARTIST": [], "SONG": [], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, SONG, ALBUM\n\nText:\nElectropop is a music genre combining elements of Electronic music and Pop music genres.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, SONG, ALBUM\n\nText:\nElectropop is a music genre combining elements of Electronic music and Pop music genres.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_417", "text": "Prior to that, in 1999, he became the undisputed light heavyweight champion by unifying the WBA, World Boxing Council, and International Boxing Federation titles.", "requested_types": ["ORGANISATION", "COUNTRY", "SONG"], "distractors": ["COUNTRY", "SONG"], "gold": {"ORGANISATION": ["WBA", "World Boxing Council", "International Boxing Federation"], "COUNTRY": [], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG\n\nText:\nPrior to that, in 1999, he became the undisputed light heavyweight champion by unifying the WBA, World Boxing Council, and International Boxing Federation titles.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG\n\nText:\nPrior to that, in 1999, he became the undisputed light heavyweight champion by unifying the WBA, World Boxing Council, and International Boxing Federation titles.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_418", "text": "After attaining mainstream success, Cobain became a devoted champion of lesser known indie bands, covering songs by The Vaselines, Meat Puppets, Wipers and Fang onstage and / or in the studio, wearing Daniel Johnston T-shirts during photo shoots, having the K Records logo tattooed on his forearm, and enlisting bands like Butthole Surfers, Shonen Knife, Chokebore and Half Japanese along for the In Utero tour in late 1993 and early 1994.", "requested_types": ["AWARD", "MUSICAL_ARTIST", "ORGANISATION", "MISC", "BAND", "MUSIC_GENRE", "ALBUM"], "distractors": ["MUSIC_GENRE", "AWARD", "ORGANISATION"], "gold": {"AWARD": [], "MUSICAL_ARTIST": ["Cobain", "Daniel Johnston"], "ORGANISATION": [], "MISC": ["K Records"], "BAND": ["The Vaselines", "Meat Puppets", "Wipers", "Fang", "Butthole Surfers", "Shonen Knife", "Chokebore", "Half Japanese"], "MUSIC_GENRE": [], "ALBUM": ["In Utero"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ORGANISATION, MISC, BAND, MUSIC_GENRE, ALBUM\n\nText:\nAfter attaining mainstream success, Cobain became a devoted champion of lesser known indie bands, covering songs by The Vaselines, Meat Puppets, Wipers and Fang onstage and / or in the studio, wearing Daniel Johnston T-shirts during photo shoots, having the K Records logo tattooed on his forearm, and enlisting bands like Butthole Surfers, Shonen Knife, Chokebore and Half Japanese along for the In Utero tour in late 1993 and early 1994.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ORGANISATION, MISC, BAND, MUSIC_GENRE, ALBUM\n\nText:\nAfter attaining mainstream success, Cobain became a devoted champion of lesser known indie bands, covering songs by The Vaselines, Meat Puppets, Wipers and Fang onstage and / or in the studio, wearing Daniel Johnston T-shirts during photo shoots, having the K Records logo tattooed on his forearm, and enlisting bands like Butthole Surfers, Shonen Knife, Chokebore and Half Japanese along for the In Utero tour in late 1993 and early 1994.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_419", "text": "Grammer has won five Primetime Emmy Awards, three Golden Globe Awards, one Daytime Emmy Award and one Tony Award; additionally, he has also worked as a television producer, director, and writer.", "requested_types": ["AWARD", "MISC", "PERSON", "MUSICAL_ARTIST", "ALBUM"], "distractors": ["ALBUM", "MISC", "PERSON"], "gold": {"AWARD": ["Primetime Emmy Awards", "Golden Globe Awards", "Daytime Emmy Award", "Tony Award"], "MISC": [], "PERSON": [], "MUSICAL_ARTIST": ["Grammer"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, MUSICAL_ARTIST, ALBUM\n\nText:\nGrammer has won five Primetime Emmy Awards, three Golden Globe Awards, one Daytime Emmy Award and one Tony Award; additionally, he has also worked as a television producer, director, and writer.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, MUSICAL_ARTIST, ALBUM\n\nText:\nGrammer has won five Primetime Emmy Awards, three Golden Globe Awards, one Daytime Emmy Award and one Tony Award; additionally, he has also worked as a television producer, director, and writer.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_420", "text": "He has won four Academy Awards, one for Academy Award for Best Director, and three for Academy Award for Best Original Screenplay.", "requested_types": ["AWARD", "BAND"], "distractors": ["BAND"], "gold": {"AWARD": ["Academy Awards", "Academy Award for Best Director", "Academy Award for Best Original Screenplay"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nHe has won four Academy Awards, one for Academy Award for Best Director, and three for Academy Award for Best Original Screenplay.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nHe has won four Academy Awards, one for Academy Award for Best Director, and three for Academy Award for Best Original Screenplay.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_421", "text": "Swamp blues has a laid-back, slow tempo, and generally is a more rhythmic variation of Louisiana blues, incorporating influences from New Orleans blues, zydeco, soul music and Cajun music.", "requested_types": ["MUSIC_GENRE", "MISC", "EVENT"], "distractors": ["EVENT", "MISC"], "gold": {"MUSIC_GENRE": ["Swamp blues", "Louisiana blues", "New Orleans blues", "zydeco", "soul music", "Cajun music"], "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: MUSIC_GENRE, MISC, EVENT\n\nText:\nSwamp blues has a laid-back, slow tempo, and generally is a more rhythmic variation of Louisiana blues, incorporating influences from New Orleans blues, zydeco, soul music and Cajun music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MISC, EVENT\n\nText:\nSwamp blues has a laid-back, slow tempo, and generally is a more rhythmic variation of Louisiana blues, incorporating influences from New Orleans blues, zydeco, soul music and Cajun music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_422", "text": "The groups released albums such as Maggot Brain (1971), Mothership Connection (1975), and One Nation Under a Groove (1978) to critical praise, and scored charting hits with singles such as Give Up the Funk (1976), One Nation Under a Groove (1978), and Flash Light (1978).", "requested_types": ["BAND", "ALBUM", "MUSICAL_ARTIST", "SONG", "MISC"], "distractors": ["BAND", "MUSICAL_ARTIST", "MISC"], "gold": {"BAND": [], "ALBUM": ["Maggot Brain", "Mothership Connection", "One Nation Under a Groove"], "MUSICAL_ARTIST": [], "SONG": ["Give Up the Funk", "One Nation Under a Groove", "Flash Light"], "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: BAND, ALBUM, MUSICAL_ARTIST, SONG, MISC\n\nText:\nThe groups released albums such as Maggot Brain (1971), Mothership Connection (1975), and One Nation Under a Groove (1978) to critical praise, and scored charting hits with singles such as Give Up the Funk (1976), One Nation Under a Groove (1978), and Flash Light (1978).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, ALBUM, MUSICAL_ARTIST, SONG, MISC\n\nText:\nThe groups released albums such as Maggot Brain (1971), Mothership Connection (1975), and One Nation Under a Groove (1978) to critical praise, and scored charting hits with singles such as Give Up the Funk (1976), One Nation Under a Groove (1978), and Flash Light (1978).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_423", "text": "The musical was nominated for three Tony Awards: Tony Award for Best Musical, Tony Award for Best Scenic Design, and Tony Award for Best Lighting Design, winning the latter two.", "requested_types": ["AWARD", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_ARTIST"], "gold": {"AWARD": ["Tony Awards", "Tony Award for Best Musical", "Tony Award for Best Scenic Design", "Tony Award for Best Lighting Design"], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST\n\nText:\nThe musical was nominated for three Tony Awards: Tony Award for Best Musical, Tony Award for Best Scenic Design, and Tony Award for Best Lighting Design, winning the latter two.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST\n\nText:\nThe musical was nominated for three Tony Awards: Tony Award for Best Musical, Tony Award for Best Scenic Design, and Tony Award for Best Lighting Design, winning the latter two.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_424", "text": "Rain Man won four Academy Awards, including Academy Award for Best Picture, Academy Award for Best Actor for Hoffman, and Academy Award for Best Director for Barry Levinson.", "requested_types": ["ORGANISATION", "AWARD", "MISC", "ALBUM", "PERSON"], "distractors": ["ORGANISATION", "ALBUM"], "gold": {"ORGANISATION": [], "AWARD": ["Academy Awards", "Academy Award for Best Picture", "Academy Award for Best Actor", "Academy Award for Best Director"], "MISC": ["Rain Man"], "ALBUM": [], "PERSON": ["Hoffman", "Barry Levinson"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, ALBUM, PERSON\n\nText:\nRain Man won four Academy Awards, including Academy Award for Best Picture, Academy Award for Best Actor for Hoffman, and Academy Award for Best Director for Barry Levinson.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, ALBUM, PERSON\n\nText:\nRain Man won four Academy Awards, including Academy Award for Best Picture, Academy Award for Best Actor for Hoffman, and Academy Award for Best Director for Barry Levinson.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_425", "text": "For his work in the Western film Unforgiven (1992) and the sports drama Million Dollar Baby (2004), Eastwood won Academy Awards for Academy Award for Best Director and Academy Award for Best Picture, as well as receiving nominations for Academy Award for Best Actor.", "requested_types": ["AWARD", "PERSON", "MISC", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_ARTIST"], "gold": {"AWARD": ["Academy Awards", "Academy Award for Best Director", "Academy Award for Best Picture", "Academy Award for Best Actor"], "PERSON": ["Eastwood"], "MISC": ["Western", "Unforgiven", "Million Dollar Baby"], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, MUSICAL_ARTIST\n\nText:\nFor his work in the Western film Unforgiven (1992) and the sports drama Million Dollar Baby (2004), Eastwood won Academy Awards for Academy Award for Best Director and Academy Award for Best Picture, as well as receiving nominations for Academy Award for Best Actor.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MISC, MUSICAL_ARTIST\n\nText:\nFor his work in the Western film Unforgiven (1992) and the sports drama Million Dollar Baby (2004), Eastwood won Academy Awards for Academy Award for Best Director and Academy Award for Best Picture, as well as receiving nominations for Academy Award for Best Actor.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_426", "text": "The 1970 releases by Black Sabbath (Black Sabbath and Paranoid) and Deep Purple (In Rock) were crucial in this regard.", "requested_types": ["MISC", "BAND", "COUNTRY", "ALBUM", "MUSICAL_ARTIST"], "distractors": ["MISC", "MUSICAL_ARTIST", "COUNTRY"], "gold": {"MISC": [], "BAND": ["Black Sabbath", "Deep Purple"], "COUNTRY": [], "ALBUM": ["Black Sabbath", "Paranoid", "In Rock"], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, COUNTRY, ALBUM, MUSICAL_ARTIST\n\nText:\nThe 1970 releases by Black Sabbath (Black Sabbath and Paranoid) and Deep Purple (In Rock) were crucial in this regard.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, COUNTRY, ALBUM, MUSICAL_ARTIST\n\nText:\nThe 1970 releases by Black Sabbath (Black Sabbath and Paranoid) and Deep Purple (In Rock) were crucial in this regard.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_427", "text": "At the height of their fame during the 1970s, the band popularized the Southern rock genre with songs such as Sweet Home Alabama and Free Bird.", "requested_types": ["MISC", "SONG", "MUSIC_GENRE", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_ARTIST", "MISC"], "gold": {"MISC": [], "SONG": ["Sweet Home Alabama", "Free Bird"], "MUSIC_GENRE": ["Southern rock"], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nAt the height of their fame during the 1970s, the band popularized the Southern rock genre with songs such as Sweet Home Alabama and Free Bird.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSIC_GENRE, MUSICAL_ARTIST\n\nText:\nAt the height of their fame during the 1970s, the band popularized the Southern rock genre with songs such as Sweet Home Alabama and Free Bird.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_428", "text": "Others, including San Francisco Bay Area's Testament and Exodus, New Jersey's Overkill, and Brazil's Sepultura and Sarcófago, also had a significant impact.", "requested_types": ["COUNTRY", "LOCATION", "MUSICAL_INSTRUMENT", "BAND"], "distractors": ["COUNTRY", "MUSICAL_INSTRUMENT"], "gold": {"COUNTRY": [], "LOCATION": ["San Francisco Bay Area", "New Jersey", "Brazil"], "MUSICAL_INSTRUMENT": [], "BAND": ["Testament", "Exodus", "Overkill", "Sepultura", "Sarcófago"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, BAND\n\nText:\nOthers, including San Francisco Bay Area's Testament and Exodus, New Jersey's Overkill, and Brazil's Sepultura and Sarcófago, also had a significant impact.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_INSTRUMENT, BAND\n\nText:\nOthers, including San Francisco Bay Area's Testament and Exodus, New Jersey's Overkill, and Brazil's Sepultura and Sarcófago, also had a significant impact.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_429", "text": "The top 20 also included LPs by The Velvet Underground, The Ramones, Pulp, Misty in Roots, Nirvana, Neil Young, Pink Floyd, The Four Brothers, Dave Clarke, Richard and Linda Thompson and The Rolling Stones.", "requested_types": ["MISC", "PERSON", "SONG", "MUSICAL_ARTIST", "BAND"], "distractors": ["PERSON", "SONG"], "gold": {"MISC": ["LPs"], "PERSON": [], "SONG": [], "MUSICAL_ARTIST": ["Neil Young", "Dave Clarke"], "BAND": ["The Velvet Underground", "The Ramones", "Pulp", "Misty in Roots", "Nirvana", "Pink Floyd", "The Four Brothers", "Richard and Linda Thompson", "The Rolling Stones"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST, BAND\n\nText:\nThe top 20 also included LPs by The Velvet Underground, The Ramones, Pulp, Misty in Roots, Nirvana, Neil Young, Pink Floyd, The Four Brothers, Dave Clarke, Richard and Linda Thompson and The Rolling Stones.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, MUSICAL_ARTIST, BAND\n\nText:\nThe top 20 also included LPs by The Velvet Underground, The Ramones, Pulp, Misty in Roots, Nirvana, Neil Young, Pink Floyd, The Four Brothers, Dave Clarke, Richard and Linda Thompson and The Rolling Stones.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_430", "text": "Nitin Mishra came up with contemporary storylines like 'Hum Honge Kamyaab' in 2010, which was his take on the controversies attached with the 2010 Commonwealth Games conducted in India in 2010.", "requested_types": ["COUNTRY", "EVENT", "SONG", "PERSON", "MUSICAL_ARTIST"], "distractors": ["PERSON"], "gold": {"COUNTRY": ["India"], "EVENT": ["2010 Commonwealth Games"], "SONG": ["Hum Honge Kamyaab"], "PERSON": [], "MUSICAL_ARTIST": ["Nitin Mishra"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, PERSON, MUSICAL_ARTIST\n\nText:\nNitin Mishra came up with contemporary storylines like 'Hum Honge Kamyaab' in 2010, which was his take on the controversies attached with the 2010 Commonwealth Games conducted in India in 2010.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, SONG, PERSON, MUSICAL_ARTIST\n\nText:\nNitin Mishra came up with contemporary storylines like 'Hum Honge Kamyaab' in 2010, which was his take on the controversies attached with the 2010 Commonwealth Games conducted in India in 2010.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_431", "text": "... Baby One More Time earned Spears Grammy Award nominations for Grammy Award for Best New Artist and Grammy Award for Best Female Pop Vocal Performance for... Baby One More Time at the 42nd Annual Grammy Awards in 2000.", "requested_types": ["SONG", "MUSICAL_ARTIST", "AWARD"], "distractors": ["MUSICAL_ARTIST"], "gold": {"SONG": ["Baby One More Time"], "MUSICAL_ARTIST": [], "AWARD": ["Grammy Award", "Grammy Award for Best New Artist", "Grammy Award for Best Female Pop Vocal Performance", "42nd Annual Grammy Awards"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, AWARD\n\nText:\n... Baby One More Time earned Spears Grammy Award nominations for Grammy Award for Best New Artist and Grammy Award for Best Female Pop Vocal Performance for... Baby One More Time at the 42nd Annual Grammy Awards in 2000.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, AWARD\n\nText:\n... Baby One More Time earned Spears Grammy Award nominations for Grammy Award for Best New Artist and Grammy Award for Best Female Pop Vocal Performance for... Baby One More Time at the 42nd Annual Grammy Awards in 2000.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_432", "text": "The current general music director of the Berlin State Opera and the Staatskapelle Berlin, Barenboim previously served as Music Director of the Chicago Symphony Orchestra, the Orchestre de Paris and La Scala in Milan.", "requested_types": ["MUSICAL_ARTIST", "MISC", "MUSICAL_INSTRUMENT", "LOCATION", "BAND"], "distractors": ["MISC", "MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_ARTIST": ["Barenboim"], "MISC": [], "MUSICAL_INSTRUMENT": [], "LOCATION": ["Berlin State Opera", "La Scala", "Milan"], "BAND": ["Staatskapelle Berlin", "Chicago Symphony Orchestra", "Orchestre de Paris"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MISC, MUSICAL_INSTRUMENT, LOCATION, BAND\n\nText:\nThe current general music director of the Berlin State Opera and the Staatskapelle Berlin, Barenboim previously served as Music Director of the Chicago Symphony Orchestra, the Orchestre de Paris and La Scala in Milan.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MISC, MUSICAL_INSTRUMENT, LOCATION, BAND\n\nText:\nThe current general music director of the Berlin State Opera and the Staatskapelle Berlin, Barenboim previously served as Music Director of the Chicago Symphony Orchestra, the Orchestre de Paris and La Scala in Milan.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_433", "text": "Building on the distorted electric guitar sound of early records, his 1958 instrumental hit Rumble by Link Wray & His Ray Men popularized the power chord, the major modus operandi of modern Rock music guitarists, facilitating the emergence of Punk rock and Hard rock.", "requested_types": ["MUSICAL_INSTRUMENT", "EVENT", "ALBUM", "MUSIC_GENRE", "BAND", "MISC"], "distractors": ["EVENT"], "gold": {"MUSICAL_INSTRUMENT": ["electric guitar"], "EVENT": [], "ALBUM": ["Rumble"], "MUSIC_GENRE": ["Rock music", "Punk rock", "Hard rock"], "BAND": ["Link Wray & His Ray Men"], "MISC": ["power chord", "modus operandi"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, EVENT, ALBUM, MUSIC_GENRE, BAND, MISC\n\nText:\nBuilding on the distorted electric guitar sound of early records, his 1958 instrumental hit Rumble by Link Wray & His Ray Men popularized the power chord, the major modus operandi of modern Rock music guitarists, facilitating the emergence of Punk rock and Hard rock.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, EVENT, ALBUM, MUSIC_GENRE, BAND, MISC\n\nText:\nBuilding on the distorted electric guitar sound of early records, his 1958 instrumental hit Rumble by Link Wray & His Ray Men popularized the power chord, the major modus operandi of modern Rock music guitarists, facilitating the emergence of Punk rock and Hard rock.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_434", "text": "Furthermore, he wrote the music for the Sestriere FIS Alpine World Ski Championships 1997 opening ceremony of 1997 including the official anthem, incidental music for stage productions, and a film score.", "requested_types": ["MISC", "LOCATION", "EVENT", "BAND"], "distractors": ["MISC", "BAND"], "gold": {"MISC": [], "LOCATION": ["Sestriere"], "EVENT": ["FIS Alpine World Ski Championships 1997"], "BAND": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nFurthermore, he wrote the music for the Sestriere FIS Alpine World Ski Championships 1997 opening ceremony of 1997 including the official anthem, incidental music for stage productions, and a film score.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND\n\nText:\nFurthermore, he wrote the music for the Sestriere FIS Alpine World Ski Championships 1997 opening ceremony of 1997 including the official anthem, incidental music for stage productions, and a film score.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_435", "text": "format = Country music, Bluegrass music, Gospel music |", "requested_types": ["MUSICAL_INSTRUMENT", "MUSIC_GENRE"], "distractors": ["MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": ["Country music", "Bluegrass music", "Gospel music"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSIC_GENRE\n\nText:\nformat = Country music, Bluegrass music, Gospel music |\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, MUSIC_GENRE\n\nText:\nformat = Country music, Bluegrass music, Gospel music |\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_436", "text": "Foreman wrote Melbourne Girl for the Closing Ceremony of the 2002 Commonwealth Games in Manchester, UK which was performed by Vanessa Amorosi.", "requested_types": ["SONG", "BAND", "MUSICAL_ARTIST", "COUNTRY", "LOCATION", "MISC", "EVENT", "ORGANISATION"], "distractors": ["MISC", "BAND", "ORGANISATION"], "gold": {"SONG": ["Melbourne Girl"], "BAND": [], "MUSICAL_ARTIST": ["Vanessa Amorosi"], "COUNTRY": ["UK"], "LOCATION": ["Manchester"], "MISC": [], "EVENT": ["2002 Commonwealth Games"], "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: SONG, BAND, MUSICAL_ARTIST, COUNTRY, LOCATION, MISC, EVENT, ORGANISATION\n\nText:\nForeman wrote Melbourne Girl for the Closing Ceremony of the 2002 Commonwealth Games in Manchester, UK which was performed by Vanessa Amorosi.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, BAND, MUSICAL_ARTIST, COUNTRY, LOCATION, MISC, EVENT, ORGANISATION\n\nText:\nForeman wrote Melbourne Girl for the Closing Ceremony of the 2002 Commonwealth Games in Manchester, UK which was performed by Vanessa Amorosi.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_437", "text": "In addition, McBride has the Country Music Association 's Female Vocalist of the Year award four times (tied with Reba McEntire for the second-most wins) and the Academy of Country Music' s Top Female Vocalist award three times.", "requested_types": ["COUNTRY", "PERSON", "MUSICAL_ARTIST", "ORGANISATION", "ALBUM", "AWARD"], "distractors": ["ALBUM", "PERSON", "COUNTRY"], "gold": {"COUNTRY": [], "PERSON": [], "MUSICAL_ARTIST": ["McBride", "Reba McEntire"], "ORGANISATION": ["Country Music Association", "Academy of Country Music"], "ALBUM": [], "AWARD": ["Female Vocalist of the Year award", "Top Female Vocalist 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, PERSON, MUSICAL_ARTIST, ORGANISATION, ALBUM, AWARD\n\nText:\nIn addition, McBride has the Country Music Association 's Female Vocalist of the Year award four times (tied with Reba McEntire for the second-most wins) and the Academy of Country Music' s Top Female Vocalist award three times.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, PERSON, MUSICAL_ARTIST, ORGANISATION, ALBUM, AWARD\n\nText:\nIn addition, McBride has the Country Music Association 's Female Vocalist of the Year award four times (tied with Reba McEntire for the second-most wins) and the Academy of Country Music' s Top Female Vocalist award three times.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_438", "text": "Boogie Down Productions was a Hip hop music group, originally composed of KRS-One, D-Nice, and DJ Scott La Rock.", "requested_types": ["MUSIC_GENRE", "MUSICAL_ARTIST", "MISC", "AWARD", "BAND"], "distractors": ["AWARD"], "gold": {"MUSIC_GENRE": ["Hip hop music"], "MUSICAL_ARTIST": ["KRS-One", "D-Nice", "Scott La Rock"], "MISC": ["DJ"], "AWARD": [], "BAND": ["Boogie Down Productions"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, MISC, AWARD, BAND\n\nText:\nBoogie Down Productions was a Hip hop music group, originally composed of KRS-One, D-Nice, and DJ Scott La Rock.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, MUSICAL_ARTIST, MISC, AWARD, BAND\n\nText:\nBoogie Down Productions was a Hip hop music group, originally composed of KRS-One, D-Nice, and DJ Scott La Rock.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_439", "text": "The track was used for the entrance of Team Scotland during the opening ceremony of the 2014 Commonwealth Games at Celtic Park in Glasgow.", "requested_types": ["MUSIC_GENRE", "LOCATION", "MUSICAL_INSTRUMENT", "EVENT", "ALBUM", "ORGANISATION"], "distractors": ["ALBUM", "MUSICAL_INSTRUMENT", "MUSIC_GENRE"], "gold": {"MUSIC_GENRE": [], "LOCATION": ["Celtic Park", "Glasgow"], "MUSICAL_INSTRUMENT": [], "EVENT": ["2014 Commonwealth Games"], "ALBUM": [], "ORGANISATION": ["Team Scotland"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, LOCATION, MUSICAL_INSTRUMENT, EVENT, ALBUM, ORGANISATION\n\nText:\nThe track was used for the entrance of Team Scotland during the opening ceremony of the 2014 Commonwealth Games at Celtic Park in Glasgow.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, LOCATION, MUSICAL_INSTRUMENT, EVENT, ALBUM, ORGANISATION\n\nText:\nThe track was used for the entrance of Team Scotland during the opening ceremony of the 2014 Commonwealth Games at Celtic Park in Glasgow.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_440", "text": "Parton recorded a series of bluegrass -inspired albums, beginning with The Grass Is Blue (1999), winning a Grammy Award for Best Bluegrass Album; and Little Sparrow (2001), with its cover of Collective Soul 's Shine winning a Grammy Award for Best Female Country Vocal Performance.", "requested_types": ["BAND", "MUSICAL_ARTIST", "AWARD", "PERSON", "MUSIC_GENRE", "SONG", "ALBUM"], "distractors": ["PERSON"], "gold": {"BAND": ["Collective Soul"], "MUSICAL_ARTIST": ["Parton"], "AWARD": ["Grammy Award for Best Bluegrass Album", "Grammy Award for Best Female Country Vocal Performance"], "PERSON": [], "MUSIC_GENRE": ["bluegrass"], "SONG": ["Shine"], "ALBUM": ["The Grass Is Blue", "Little Sparrow"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSICAL_ARTIST, AWARD, PERSON, MUSIC_GENRE, SONG, ALBUM\n\nText:\nParton recorded a series of bluegrass -inspired albums, beginning with The Grass Is Blue (1999), winning a Grammy Award for Best Bluegrass Album; and Little Sparrow (2001), with its cover of Collective Soul 's Shine winning a Grammy Award for Best Female Country Vocal Performance.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, MUSICAL_ARTIST, AWARD, PERSON, MUSIC_GENRE, SONG, ALBUM\n\nText:\nParton recorded a series of bluegrass -inspired albums, beginning with The Grass Is Blue (1999), winning a Grammy Award for Best Bluegrass Album; and Little Sparrow (2001), with its cover of Collective Soul 's Shine winning a Grammy Award for Best Female Country Vocal Performance.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_441", "text": "The Doors incorporated Asturias into their song Spanish Caravan; also, Iron Maiden 's Piece of Mind uses the introduction of the piece for the song bridge.", "requested_types": ["SONG", "MUSICAL_ARTIST", "MISC", "ALBUM", "BAND", "LOCATION"], "distractors": ["MISC", "MUSICAL_ARTIST"], "gold": {"SONG": ["Spanish Caravan"], "MUSICAL_ARTIST": [], "MISC": [], "ALBUM": ["Piece of Mind"], "BAND": ["The Doors", "Iron Maiden"], "LOCATION": ["Asturias"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, MISC, ALBUM, BAND, LOCATION\n\nText:\nThe Doors incorporated Asturias into their song Spanish Caravan; also, Iron Maiden 's Piece of Mind uses the introduction of the piece for the song bridge.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: SONG, MUSICAL_ARTIST, MISC, ALBUM, BAND, LOCATION\n\nText:\nThe Doors incorporated Asturias into their song Spanish Caravan; also, Iron Maiden 's Piece of Mind uses the introduction of the piece for the song bridge.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_442", "text": "Boogie-woogie was pioneered by the Chicago-based Jimmy Yancey and the Boogie-Woogie Trio (Albert Ammons, Pete Johnson and Meade Lux Lewis).", "requested_types": ["MUSICAL_ARTIST", "MISC", "LOCATION", "MUSIC_GENRE", "ORGANISATION"], "distractors": ["ORGANISATION"], "gold": {"MUSICAL_ARTIST": ["Jimmy Yancey", "Albert Ammons", "Pete Johnson", "Meade Lux Lewis"], "MISC": ["Boogie-Woogie Trio"], "LOCATION": ["Chicago-based"], "MUSIC_GENRE": ["Boogie-woogie"], "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: MUSICAL_ARTIST, MISC, LOCATION, MUSIC_GENRE, ORGANISATION\n\nText:\nBoogie-woogie was pioneered by the Chicago-based Jimmy Yancey and the Boogie-Woogie Trio (Albert Ammons, Pete Johnson and Meade Lux Lewis).\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, MISC, LOCATION, MUSIC_GENRE, ORGANISATION\n\nText:\nBoogie-woogie was pioneered by the Chicago-based Jimmy Yancey and the Boogie-Woogie Trio (Albert Ammons, Pete Johnson and Meade Lux Lewis).\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_443", "text": "The music was a reaction to the 1980s flow of hip hop culture from New York City and the Los Angeles area and can be considered a third major American hip hop genre, after East Coast hip hop and West Coast hip hop.", "requested_types": ["MISC", "LOCATION", "EVENT", "MUSIC_GENRE", "ALBUM"], "distractors": ["ALBUM", "EVENT"], "gold": {"MISC": ["American"], "LOCATION": ["New York City", "Los Angeles"], "EVENT": [], "MUSIC_GENRE": ["hip hop", "East Coast hip hop", "West Coast hip hop"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, ALBUM\n\nText:\nThe music was a reaction to the 1980s flow of hip hop culture from New York City and the Los Angeles area and can be considered a third major American hip hop genre, after East Coast hip hop and West Coast hip hop.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, ALBUM\n\nText:\nThe music was a reaction to the 1980s flow of hip hop culture from New York City and the Los Angeles area and can be considered a third major American hip hop genre, after East Coast hip hop and West Coast hip hop.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_444", "text": "Hathaway currently serves on the board of the Lollipop Theatre Network and is involved with charities Creative Coalition, St. Jude Children's Research Hospital and the Human Rights Campaign.", "requested_types": ["ORGANISATION", "PERSON", "ALBUM"], "distractors": ["ALBUM"], "gold": {"ORGANISATION": ["Lollipop Theatre Network", "Creative Coalition", "St. Jude Children's Research Hospital", "Human Rights Campaign"], "PERSON": ["Hathaway"], "ALBUM": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nHathaway currently serves on the board of the Lollipop Theatre Network and is involved with charities Creative Coalition, St. Jude Children's Research Hospital and the Human Rights Campaign.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ALBUM\n\nText:\nHathaway currently serves on the board of the Lollipop Theatre Network and is involved with charities Creative Coalition, St. Jude Children's Research Hospital and the Human Rights Campaign.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_445", "text": "O'Connell co-hosted the Miss USA and Miss Universe pageants with Bob Barker from 1972 to 1980 and was nominated for an Emmy award in 1976 for her coverage of the Miss Universe pageant.", "requested_types": ["ORGANISATION", "PERSON", "BAND", "MUSICAL_ARTIST", "AWARD", "EVENT"], "distractors": ["EVENT", "BAND"], "gold": {"ORGANISATION": ["Miss USA", "Miss Universe"], "PERSON": ["Bob Barker"], "BAND": [], "MUSICAL_ARTIST": ["O'Connell"], "AWARD": ["Emmy 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: ORGANISATION, PERSON, BAND, MUSICAL_ARTIST, AWARD, EVENT\n\nText:\nO'Connell co-hosted the Miss USA and Miss Universe pageants with Bob Barker from 1972 to 1980 and was nominated for an Emmy award in 1976 for her coverage of the Miss Universe pageant.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSICAL_ARTIST, AWARD, EVENT\n\nText:\nO'Connell co-hosted the Miss USA and Miss Universe pageants with Bob Barker from 1972 to 1980 and was nominated for an Emmy award in 1976 for her coverage of the Miss Universe pageant.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_446", "text": "Sosa performed in venues such as the Lincoln Center in New York City, the Théâtre Mogador in Paris and the Sistine Chapel in Vatican City, as well as sell-out shows in New York's Carnegie Hall and the Roman Colosseum during her final decade of life.", "requested_types": ["MUSICAL_ARTIST", "LOCATION", "AWARD", "SONG", "ORGANISATION"], "distractors": ["SONG", "AWARD", "ORGANISATION"], "gold": {"MUSICAL_ARTIST": ["Sosa"], "LOCATION": ["Lincoln Center", "New York City", "Théâtre Mogador", "Paris", "Sistine Chapel", "Vatican City", "New York", "Carnegie Hall", "Roman Colosseum"], "AWARD": [], "SONG": [], "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: MUSICAL_ARTIST, LOCATION, AWARD, SONG, ORGANISATION\n\nText:\nSosa performed in venues such as the Lincoln Center in New York City, the Théâtre Mogador in Paris and the Sistine Chapel in Vatican City, as well as sell-out shows in New York's Carnegie Hall and the Roman Colosseum during her final decade of life.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, LOCATION, AWARD, SONG, ORGANISATION\n\nText:\nSosa performed in venues such as the Lincoln Center in New York City, the Théâtre Mogador in Paris and the Sistine Chapel in Vatican City, as well as sell-out shows in New York's Carnegie Hall and the Roman Colosseum during her final decade of life.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_447", "text": "From the 1970s through the 1990s, Terry performed at Carnegie Hall, Town Hall, and Lincoln Center, toured with the Newport Jazz All Stars and Jazz at the Philharmonic, and was featured with Skitch Henderson 's New York Pops Orchestra.", "requested_types": ["EVENT", "MUSICAL_ARTIST", "AWARD", "LOCATION"], "distractors": ["AWARD"], "gold": {"EVENT": ["Newport Jazz All Stars", "Jazz at the Philharmonic"], "MUSICAL_ARTIST": ["Terry", "Skitch Henderson"], "AWARD": [], "LOCATION": ["Carnegie Hall", "Town Hall", "Lincoln Center", "New York Pops Orchestra"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, AWARD, LOCATION\n\nText:\nFrom the 1970s through the 1990s, Terry performed at Carnegie Hall, Town Hall, and Lincoln Center, toured with the Newport Jazz All Stars and Jazz at the Philharmonic, and was featured with Skitch Henderson 's New York Pops Orchestra.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, AWARD, LOCATION\n\nText:\nFrom the 1970s through the 1990s, Terry performed at Carnegie Hall, Town Hall, and Lincoln Center, toured with the Newport Jazz All Stars and Jazz at the Philharmonic, and was featured with Skitch Henderson 's New York Pops Orchestra.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_448", "text": "By the early 20th century, the United States had become a major center for folk music from around the world, including polka, Ukrainian and Polish fiddling, Ashkenazi, Klezmer, and several kinds of Latin music.", "requested_types": ["MUSIC_GENRE", "COUNTRY", "EVENT", "MISC", "LOCATION"], "distractors": ["MISC", "LOCATION", "EVENT"], "gold": {"MUSIC_GENRE": ["folk music", "polka", "Ukrainian", "Polish fiddling", "Ashkenazi", "Klezmer", "Latin music"], "COUNTRY": ["United States"], "EVENT": [], "MISC": [], "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: MUSIC_GENRE, COUNTRY, EVENT, MISC, LOCATION\n\nText:\nBy the early 20th century, the United States had become a major center for folk music from around the world, including polka, Ukrainian and Polish fiddling, Ashkenazi, Klezmer, and several kinds of Latin music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, COUNTRY, EVENT, MISC, LOCATION\n\nText:\nBy the early 20th century, the United States had become a major center for folk music from around the world, including polka, Ukrainian and Polish fiddling, Ashkenazi, Klezmer, and several kinds of Latin music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_449", "text": "Although Chamberlain left DA in late 1983, he did return on a few later albums including MotorCycle (1993) (which he co-produced), BibleLand (1994) and Songs of the Heart (1995).", "requested_types": ["ALBUM", "MUSICAL_ARTIST", "MUSICAL_INSTRUMENT", "MUSIC_GENRE", "BAND"], "distractors": ["MUSICAL_INSTRUMENT", "MUSIC_GENRE"], "gold": {"ALBUM": ["MotorCycle", "BibleLand", "Songs of the Heart"], "MUSICAL_ARTIST": ["Chamberlain"], "MUSICAL_INSTRUMENT": [], "MUSIC_GENRE": [], "BAND": ["DA"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, MUSIC_GENRE, BAND\n\nText:\nAlthough Chamberlain left DA in late 1983, he did return on a few later albums including MotorCycle (1993) (which he co-produced), BibleLand (1994) and Songs of the Heart (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: ALBUM, MUSICAL_ARTIST, MUSICAL_INSTRUMENT, MUSIC_GENRE, BAND\n\nText:\nAlthough Chamberlain left DA in late 1983, he did return on a few later albums including MotorCycle (1993) (which he co-produced), BibleLand (1994) and Songs of the Heart (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_music_450", "text": "Since 2000, France, Germany, Spain and United Kingdom have automatically qualified for the final, regardless of their positions on the scoreboard in previous contests, as they are the four biggest financial contributors to the European Broadcasting Union.", "requested_types": ["COUNTRY", "LOCATION", "ORGANISATION"], "distractors": ["LOCATION"], "gold": {"COUNTRY": ["France", "Germany", "Spain", "United Kingdom"], "LOCATION": [], "ORGANISATION": ["European Broadcasting 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: COUNTRY, LOCATION, ORGANISATION\n\nText:\nSince 2000, France, Germany, Spain and United Kingdom have automatically qualified for the final, regardless of their positions on the scoreboard in previous contests, as they are the four biggest financial contributors to the European Broadcasting Union.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, ORGANISATION\n\nText:\nSince 2000, France, Germany, Spain and United Kingdom have automatically qualified for the final, regardless of their positions on the scoreboard in previous contests, as they are the four biggest financial contributors to the European Broadcasting Union.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_451", "text": "Country influences combined with Punk rock and alternative rock to forge the cowpunk scene in Southern California during the 1980s, which included bands such as The Long Ryders, Lone Justice and The Beat Farmers, as well as the established punk group X, whose music had begun to include country and rockabilly influences.", "requested_types": ["COUNTRY", "AWARD", "LOCATION", "BAND", "MUSIC_GENRE", "SONG"], "distractors": ["SONG", "COUNTRY", "AWARD"], "gold": {"COUNTRY": [], "AWARD": [], "LOCATION": ["Southern California"], "BAND": ["The Long Ryders", "Lone Justice", "The Beat Farmers", "X"], "MUSIC_GENRE": ["Punk rock", "alternative rock", "cowpunk", "punk", "country", "rockabilly"], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, BAND, MUSIC_GENRE, SONG\n\nText:\nCountry influences combined with Punk rock and alternative rock to forge the cowpunk scene in Southern California during the 1980s, which included bands such as The Long Ryders, Lone Justice and The Beat Farmers, as well as the established punk group X, whose music had begun to include country and rockabilly 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: COUNTRY, AWARD, LOCATION, BAND, MUSIC_GENRE, SONG\n\nText:\nCountry influences combined with Punk rock and alternative rock to forge the cowpunk scene in Southern California during the 1980s, which included bands such as The Long Ryders, Lone Justice and The Beat Farmers, as well as the established punk group X, whose music had begun to include country and rockabilly 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_music_452", "text": "Derived from the traditional Western, including Red Dirt, New Mexico music, Texas country music, Tejano music, and honky-tonk musical styles of the late 1950s and 1960s.", "requested_types": ["MUSIC_GENRE", "AWARD", "BAND", "MISC"], "distractors": ["AWARD", "BAND", "MISC"], "gold": {"MUSIC_GENRE": ["Western", "Red Dirt", "New Mexico music", "Texas country music", "Tejano music", "honky-tonk"], "AWARD": [], "BAND": [], "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: MUSIC_GENRE, AWARD, BAND, MISC\n\nText:\nDerived from the traditional Western, including Red Dirt, New Mexico music, Texas country music, Tejano music, and honky-tonk musical styles of the late 1950s and 1960s.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, AWARD, BAND, MISC\n\nText:\nDerived from the traditional Western, including Red Dirt, New Mexico music, Texas country music, Tejano music, and honky-tonk musical styles of the late 1950s and 1960s.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_453", "text": "Independent from the United Kingdom scene, in the late 1970s and early 1980s in California, deathrock developed as a distinct branch of American punk rock, with acts such as Christian Death and 45 Grave.", "requested_types": ["ALBUM", "ORGANISATION", "MISC", "SONG", "BAND", "LOCATION", "COUNTRY", "MUSIC_GENRE"], "distractors": ["ALBUM", "SONG", "ORGANISATION"], "gold": {"ALBUM": [], "ORGANISATION": [], "MISC": ["American"], "SONG": [], "BAND": ["Christian Death", "45 Grave"], "LOCATION": ["California"], "COUNTRY": ["United Kingdom"], "MUSIC_GENRE": ["deathrock", "punk rock"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, ORGANISATION, MISC, SONG, BAND, LOCATION, COUNTRY, MUSIC_GENRE\n\nText:\nIndependent from the United Kingdom scene, in the late 1970s and early 1980s in California, deathrock developed as a distinct branch of American punk rock, with acts such as Christian Death and 45 Grave.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, ORGANISATION, MISC, SONG, BAND, LOCATION, COUNTRY, MUSIC_GENRE\n\nText:\nIndependent from the United Kingdom scene, in the late 1970s and early 1980s in California, deathrock developed as a distinct branch of American punk rock, with acts such as Christian Death and 45 Grave.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_454", "text": "Drummer Vinny Appice left the band following the reunion show to rejoin Ronnie James Dio's solo band, later appearing on Dio's Strange Highways and Angry Machines.", "requested_types": ["ALBUM", "MUSICAL_INSTRUMENT", "COUNTRY", "MUSICAL_ARTIST", "MUSIC_GENRE"], "distractors": ["MUSIC_GENRE", "COUNTRY", "MUSICAL_INSTRUMENT"], "gold": {"ALBUM": ["Strange Highways", "Angry Machines"], "MUSICAL_INSTRUMENT": [], "COUNTRY": [], "MUSICAL_ARTIST": ["Vinny Appice", "Ronnie James Dio", "Dio"], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_INSTRUMENT, COUNTRY, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nDrummer Vinny Appice left the band following the reunion show to rejoin Ronnie James Dio's solo band, later appearing on Dio's Strange Highways and Angry Machines.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: ALBUM, MUSICAL_INSTRUMENT, COUNTRY, MUSICAL_ARTIST, MUSIC_GENRE\n\nText:\nDrummer Vinny Appice left the band following the reunion show to rejoin Ronnie James Dio's solo band, later appearing on Dio's Strange Highways and Angry Machines.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_455", "text": "Autechre remain going strong to this day, where they continue to call Warp Records their home, having released numerous albums to critical acclaim in the years to follow, such as Confield, Draft 7.30, Untilted, Quaristice, Oversteps, Exai and elseq 1-5.", "requested_types": ["EVENT", "ORGANISATION", "MUSICAL_ARTIST", "ALBUM", "MUSICAL_INSTRUMENT", "SONG"], "distractors": ["SONG", "EVENT", "MUSICAL_INSTRUMENT"], "gold": {"EVENT": [], "ORGANISATION": ["Warp Records"], "MUSICAL_ARTIST": ["Autechre"], "ALBUM": ["Confield", "Draft 7.30", "Untilted", "Quaristice", "Oversteps", "Exai", "elseq 1-5"], "MUSICAL_INSTRUMENT": [], "SONG": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ALBUM, MUSICAL_INSTRUMENT, SONG\n\nText:\nAutechre remain going strong to this day, where they continue to call Warp Records their home, having released numerous albums to critical acclaim in the years to follow, such as Confield, Draft 7.30, Untilted, Quaristice, Oversteps, Exai and elseq 1-5.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSICAL_ARTIST, ALBUM, MUSICAL_INSTRUMENT, SONG\n\nText:\nAutechre remain going strong to this day, where they continue to call Warp Records their home, having released numerous albums to critical acclaim in the years to follow, such as Confield, Draft 7.30, Untilted, Quaristice, Oversteps, Exai and elseq 1-5.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_456", "text": "Crust punk groups, such as Antisect, Sacrilege and Anti System took some influence from early black metal bands like Venom, Hellhammer, and Celtic Frost, In addition, Norwegian band Darkthrone have incorporated crust punk traits in their more recent material.", "requested_types": ["MISC", "MUSIC_GENRE", "EVENT", "BAND"], "distractors": ["EVENT"], "gold": {"MISC": ["Norwegian band"], "MUSIC_GENRE": ["Crust punk", "black metal", "crust punk"], "EVENT": [], "BAND": ["Antisect", "Sacrilege", "Anti System", "Venom", "Hellhammer", "Celtic Frost", "Darkthrone"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, EVENT, BAND\n\nText:\nCrust punk groups, such as Antisect, Sacrilege and Anti System took some influence from early black metal bands like Venom, Hellhammer, and Celtic Frost, In addition, Norwegian band Darkthrone have incorporated crust punk traits in their more recent material.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, EVENT, BAND\n\nText:\nCrust punk groups, such as Antisect, Sacrilege and Anti System took some influence from early black metal bands like Venom, Hellhammer, and Celtic Frost, In addition, Norwegian band Darkthrone have incorporated crust punk traits in their more recent material.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_457", "text": "Madlib samples Gong's music frequently, including Eat That Phone Book Coda on his song The Further Adventures of Lord Quas, You Never Blow Yr Trip Forever on The Further Adventures of Lord Quas, and Shamal on The Further Adventures of Lord Quas.", "requested_types": ["MUSICAL_ARTIST", "SONG", "BAND", "EVENT", "ALBUM"], "distractors": ["EVENT"], "gold": {"MUSICAL_ARTIST": ["Madlib"], "SONG": ["Eat That Phone Book Coda", "You Never Blow Yr Trip Forever", "Shamal"], "BAND": ["Gong"], "EVENT": [], "ALBUM": ["The Further Adventures of Lord Quas"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, SONG, BAND, EVENT, ALBUM\n\nText:\nMadlib samples Gong's music frequently, including Eat That Phone Book Coda on his song The Further Adventures of Lord Quas, You Never Blow Yr Trip Forever on The Further Adventures of Lord Quas, and Shamal on The Further Adventures of Lord Quas.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, SONG, BAND, EVENT, ALBUM\n\nText:\nMadlib samples Gong's music frequently, including Eat That Phone Book Coda on his song The Further Adventures of Lord Quas, You Never Blow Yr Trip Forever on The Further Adventures of Lord Quas, and Shamal on The Further Adventures of Lord Quas.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_458", "text": "After releasing only one album on a major label, Black Elvis / Lost in Space, Kool Keith subsequently returned to independently releasing music, producing further efforts as a solo artist and in collaboration with groups such as Analog Brothers, Masters of Illusion, Thee Undatakerz and Project Polaroid.", "requested_types": ["MUSICAL_ARTIST", "SONG", "ALBUM", "MISC", "BAND", "MUSIC_GENRE"], "distractors": ["MUSIC_GENRE", "MISC", "SONG"], "gold": {"MUSICAL_ARTIST": ["Kool Keith"], "SONG": [], "ALBUM": ["Black Elvis / Lost in Space"], "MISC": [], "BAND": ["Analog Brothers", "Masters of Illusion", "Thee Undatakerz", "Project Polaroid"], "MUSIC_GENRE": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, SONG, ALBUM, MISC, BAND, MUSIC_GENRE\n\nText:\nAfter releasing only one album on a major label, Black Elvis / Lost in Space, Kool Keith subsequently returned to independently releasing music, producing further efforts as a solo artist and in collaboration with groups such as Analog Brothers, Masters of Illusion, Thee Undatakerz and Project Polaroid.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_ARTIST, SONG, ALBUM, MISC, BAND, MUSIC_GENRE\n\nText:\nAfter releasing only one album on a major label, Black Elvis / Lost in Space, Kool Keith subsequently returned to independently releasing music, producing further efforts as a solo artist and in collaboration with groups such as Analog Brothers, Masters of Illusion, Thee Undatakerz and Project Polaroid.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_459", "text": "Rhythms of Resistance formed as part of the UK Earth First action against the International Monetary Fund / World Bank meeting in Prague in September 2000.", "requested_types": ["BAND", "LOCATION", "ORGANISATION", "MISC"], "distractors": ["MISC"], "gold": {"BAND": ["Rhythms of Resistance"], "LOCATION": ["Prague"], "ORGANISATION": ["UK Earth First", "International Monetary Fund", "World Bank"], "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: BAND, LOCATION, ORGANISATION, MISC\n\nText:\nRhythms of Resistance formed as part of the UK Earth First action against the International Monetary Fund / World Bank meeting in Prague in September 2000.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: BAND, LOCATION, ORGANISATION, MISC\n\nText:\nRhythms of Resistance formed as part of the UK Earth First action against the International Monetary Fund / World Bank meeting in Prague in September 2000.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_460", "text": "By the Way was released in July 2002 and produced four singles; By the Way, The Zephyr Song, Can't Stop and Universally Speaking.", "requested_types": ["MUSICAL_INSTRUMENT", "LOCATION", "ALBUM", "SONG"], "distractors": ["LOCATION", "MUSICAL_INSTRUMENT"], "gold": {"MUSICAL_INSTRUMENT": [], "LOCATION": [], "ALBUM": ["By the Way"], "SONG": ["By the Way", "The Zephyr Song", "Can't Stop", "Universally Speaking"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, LOCATION, ALBUM, SONG\n\nText:\nBy the Way was released in July 2002 and produced four singles; By the Way, The Zephyr Song, Can't Stop and Universally Speaking.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSICAL_INSTRUMENT, LOCATION, ALBUM, SONG\n\nText:\nBy the Way was released in July 2002 and produced four singles; By the Way, The Zephyr Song, Can't Stop and Universally Speaking.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_461", "text": "Due to its highly syncretic nature, Latin American music encompasses a wide variety of styles, including influential genres such as cumbia, bachata, bossa nova, Merengue music, Cuban rumba, Salsa music, samba, son, and Tango music.", "requested_types": ["MISC", "MUSIC_GENRE", "ORGANISATION", "COUNTRY", "EVENT"], "distractors": ["EVENT", "COUNTRY", "ORGANISATION"], "gold": {"MISC": ["Latin American"], "MUSIC_GENRE": ["cumbia", "bachata", "bossa nova", "Merengue music", "Cuban rumba", "Salsa music", "samba", "son", "Tango music"], "ORGANISATION": [], "COUNTRY": [], "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, MUSIC_GENRE, ORGANISATION, COUNTRY, EVENT\n\nText:\nDue to its highly syncretic nature, Latin American music encompasses a wide variety of styles, including influential genres such as cumbia, bachata, bossa nova, Merengue music, Cuban rumba, Salsa music, samba, son, and Tango music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, MUSIC_GENRE, ORGANISATION, COUNTRY, EVENT\n\nText:\nDue to its highly syncretic nature, Latin American music encompasses a wide variety of styles, including influential genres such as cumbia, bachata, bossa nova, Merengue music, Cuban rumba, Salsa music, samba, son, and Tango music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_462", "text": "Crosby gained valuable experience on tour for a year with Whiteman and performing and recording with Bix Beiderbecke, Jack Teagarden, Tommy Dorsey, Jimmy Dorsey, Eddie Lang, and Hoagy Carmichael.", "requested_types": ["EVENT", "BAND", "MUSICAL_ARTIST"], "distractors": ["EVENT", "BAND"], "gold": {"EVENT": [], "BAND": [], "MUSICAL_ARTIST": ["Crosby", "Whiteman", "Bix Beiderbecke", "Jack Teagarden", "Tommy Dorsey", "Jimmy Dorsey", "Eddie Lang", "Hoagy Carmichael"]}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSICAL_ARTIST\n\nText:\nCrosby gained valuable experience on tour for a year with Whiteman and performing and recording with Bix Beiderbecke, Jack Teagarden, Tommy Dorsey, Jimmy Dorsey, Eddie Lang, and Hoagy Carmichael.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, BAND, MUSICAL_ARTIST\n\nText:\nCrosby gained valuable experience on tour for a year with Whiteman and performing and recording with Bix Beiderbecke, Jack Teagarden, Tommy Dorsey, Jimmy Dorsey, Eddie Lang, and Hoagy Carmichael.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_463", "text": "Today there are Celtic-influenced subgenres of virtually every type of popular music including electronica, Celtic rock, Celtic metal, Celtic punk, Hip hop music, reggae, New-age music, Latin, Andean and Pop music.", "requested_types": ["MUSIC_GENRE", "PERSON", "MISC", "MUSICAL_ARTIST"], "distractors": ["MUSICAL_ARTIST", "PERSON"], "gold": {"MUSIC_GENRE": ["popular music", "electronica", "Celtic rock", "Celtic metal", "Celtic punk", "Hip hop music", "reggae", "New-age music", "Latin", "Andean", "Pop music"], "PERSON": [], "MISC": ["Celtic-influenced subgenres"], "MUSICAL_ARTIST": []}, "free_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, PERSON, MISC, MUSICAL_ARTIST\n\nText:\nToday there are Celtic-influenced subgenres of virtually every type of popular music including electronica, Celtic rock, Celtic metal, Celtic punk, Hip hop music, reggae, New-age music, Latin, Andean and Pop music.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and whose values are lists of mentions copied verbatim from the text. If a type has no mentions, use an empty list.\n\nEntity types: MUSIC_GENRE, PERSON, MISC, MUSICAL_ARTIST\n\nText:\nToday there are Celtic-influenced subgenres of virtually every type of popular music including electronica, Celtic rock, Celtic metal, Celtic punk, Hip hop music, reggae, New-age music, Latin, Andean and Pop music.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}, {"id": "crossner_music_464", "text": "But in 2009, Cape Town Opera 's production, set in 1970s South Africa and inspired by life in Soweto, toured Britain, opening at the Wales Millennium Centre in Cardiff and going on to the Royal Festival Hall in London and Edinburgh Festival Theatre.", "requested_types": ["ORGANISATION", "COUNTRY", "LOCATION", "MISC"], "distractors": ["MISC"], "gold": {"ORGANISATION": ["Cape Town Opera"], "COUNTRY": ["South Africa"], "LOCATION": ["Soweto", "Britain", "Wales Millennium Centre", "Cardiff", "Royal Festival Hall", "London", "Edinburgh Festival Theatre"], "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, COUNTRY, LOCATION, MISC\n\nText:\nBut in 2009, Cape Town Opera 's production, set in 1970s South Africa and inspired by life in Soweto, toured Britain, opening at the Wales Millennium Centre in Cardiff and going on to the Royal Festival Hall in London and Edinburgh Festival Theatre.\n\n", "structured_prompt": "Extract named entities from the text for the given entity types.\nReturn a JSON object whose keys are exactly the given entity types and 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, LOCATION, MISC\n\nText:\nBut in 2009, Cape Town Opera 's production, set in 1970s South Africa and inspired by life in Soweto, toured Britain, opening at the Wales Millennium Centre in Cardiff and going on to the Royal Festival Hall in London and Edinburgh Festival Theatre.\n\n\nUse exactly these fields in your thinking, in this order. Write each field on one line.\nTYPES: Which of the requested types occur in the text, and which appear to be absent.\nMENTIONS: The mentions to extract for the types that occur, and any plausible ones you rejected.\nEDGE: Ambiguous boundaries or type assignments, and how you resolved them.\n\n"}]