{"id":"https://openalex.org/W4400949264","doi":"https://doi.org/10.1038/s41586-024-07566-y","title":"AI models collapse when trained on recursively generated data","display_name":"AI models collapse when trained on recursively generated data","publication_year":2024,"publication_date":"2024-07-24","ids":{"openalex":"https://openalex.org/W4400949264","doi":"https://doi.org/10.1038/s41586-024-07566-y","pmid":"https://pubmed.ncbi.nlm.nih.gov/39048682"},"language":"en","primary_location":{"id":"doi:10.1038/s41586-024-07566-y","is_oa":true,"landing_page_url":"https://doi.org/10.1038/s41586-024-07566-y","pdf_url":"https://www.nature.com/articles/s41586-024-07566-y.pdf","source":{"id":"https://openalex.org/S137773608","display_name":"Nature","issn_l":"0028-0836","issn":["0028-0836","1476-4687"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319908","host_organization_name":"Nature Portfolio","host_organization_lineage":["https://openalex.org/P4310319908","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Nature Portfolio","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Nature","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://www.nature.com/articles/s41586-024-07566-y.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069844959","display_name":"Ilia Shumailov","orcid":"https://orcid.org/0000-0003-3100-0727"},"institutions":[{"id":"https://openalex.org/I40120149","display_name":"University of Oxford","ror":"https://ror.org/052gg0110","country_code":"GB","type":"education","lineage":["https://openalex.org/I40120149"]}],"countries":["GB"],"is_corresponding":true,"raw_author_name":"Ilia Shumailov","raw_affiliation_strings":["OATML, Department of Computer Science, University of Oxford, Oxford, UK. ilia.shumailov@chch.ox.ac.uk"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"OATML, Department of Computer Science, University of Oxford, Oxford, UK. ilia.shumailov@chch.ox.ac.uk","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087214686","display_name":"Zakhar Shumaylov","orcid":"https://orcid.org/0000-0001-7087-4393"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Zakhar Shumaylov","raw_affiliation_strings":["Department of Applied Mathematics and Theoretical Physics, University of Cambridge, Cambridge, UK. zs334@cam.ac.uk"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Applied Mathematics and Theoretical Physics, University of Cambridge, Cambridge, UK. zs334@cam.ac.uk","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101093262","display_name":"Yiren Zhao","orcid":"https://orcid.org/0000-0002-3727-7463"},"institutions":[{"id":"https://openalex.org/I47508984","display_name":"Imperial College London","ror":"https://ror.org/041kmwe10","country_code":"GB","type":"education","lineage":["https://openalex.org/I47508984"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yiren Zhao","raw_affiliation_strings":["Department of Electrical and Electronic Engineering, Imperial College London, London, UK"],"raw_orcid":"https://orcid.org/0000-0002-3727-7463","affiliations":[{"raw_affiliation_string":"Department of Electrical and Electronic Engineering, Imperial College London, London, UK","institution_ids":["https://openalex.org/I47508984"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018809423","display_name":"Nicolas Papernot","orcid":"https://orcid.org/0000-0001-5078-7233"},"institutions":[{"id":"https://openalex.org/I185261750","display_name":"University of Toronto","ror":"https://ror.org/03dbr7087","country_code":"CA","type":"education","lineage":["https://openalex.org/I185261750"]},{"id":"https://openalex.org/I4210127509","display_name":"Vector Institute","ror":"https://ror.org/03kqdja62","country_code":"CA","type":"facility","lineage":["https://openalex.org/I4210127509"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Nicolas Papernot","raw_affiliation_strings":["University of Toronto, Toronto, Ontario, Canada","Vector Institute, Toronto, Ontario, Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Toronto, Toronto, Ontario, Canada","institution_ids":["https://openalex.org/I185261750"]},{"raw_affiliation_string":"Vector Institute, Toronto, Ontario, Canada","institution_ids":["https://openalex.org/I4210127509"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046983053","display_name":"Ross Anderson","orcid":"https://orcid.org/0000-0001-8697-5682"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]},{"id":"https://openalex.org/I98677209","display_name":"University of Edinburgh","ror":"https://ror.org/01nrxwf90","country_code":"GB","type":"education","lineage":["https://openalex.org/I98677209"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Ross Anderson","raw_affiliation_strings":["Department of Computer Science and Technology, University of Cambridge, Cambridge, UK","School of Informatics, University of Edinburgh, Edinburgh, UK"],"raw_orcid":"https://orcid.org/0000-0001-8697-5682","affiliations":[{"raw_affiliation_string":"Department of Computer Science and Technology, University of Cambridge, Cambridge, UK","institution_ids":["https://openalex.org/I241749"]},{"raw_affiliation_string":"School of Informatics, University of Edinburgh, Edinburgh, UK","institution_ids":["https://openalex.org/I98677209"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5029186201","display_name":"Yarin Gal","orcid":"https://orcid.org/0000-0002-2733-2078"},"institutions":[{"id":"https://openalex.org/I4210146410","display_name":"Science Oxford","ror":"https://ror.org/04j8yhy50","country_code":"GB","type":"nonprofit","lineage":["https://openalex.org/I4210146410"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yarin Gal","raw_affiliation_strings":["OATML, Department of Computer Science, University of Oxford, Oxford, UK. yarin@cs.ox.ac.uk"],"raw_orcid":"https://orcid.org/0000-0002-2733-2078","affiliations":[{"raw_affiliation_string":"OATML, Department of Computer Science, University of Oxford, Oxford, UK. yarin@cs.ox.ac.uk","institution_ids":["https://openalex.org/I4210146410"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5069844959"],"corresponding_institution_ids":["https://openalex.org/I40120149"],"apc_list":{"value":9750,"currency":"EUR","value_usd":11690},"apc_paid":{"value":9750,"currency":"EUR","value_usd":11690},"fwci":173.1871,"has_fulltext":true,"cited_by_count":560,"citation_normalized_percentile":{"value":0.99989623,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":"631","issue":"8022","first_page":"755","last_page":"759"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9975000023841858,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9975000023841858,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.986299991607666,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9772999882698059,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.7557709217071533},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.6722735166549683},{"id":"https://openalex.org/keywords/intuition","display_name":"Intuition","score":0.636785089969635},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.54905766248703},{"id":"https://openalex.org/keywords/variety","display_name":"Variety (cybernetics)","score":0.4976237118244171},{"id":"https://openalex.org/keywords/the-internet","display_name":"The Internet","score":0.486868292093277},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4264361262321472},{"id":"https://openalex.org/keywords/data-science","display_name":"Data science","score":0.330757200717926},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.22296109795570374},{"id":"https://openalex.org/keywords/cognitive-science","display_name":"Cognitive science","score":0.20436376333236694},{"id":"https://openalex.org/keywords/world-wide-web","display_name":"World Wide Web","score":0.1717396080493927}],"concepts":[{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.7557709217071533},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.6722735166549683},{"id":"https://openalex.org/C132010649","wikidata":"https://www.wikidata.org/wiki/Q189222","display_name":"Intuition","level":2,"score":0.636785089969635},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.54905766248703},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.4976237118244171},{"id":"https://openalex.org/C110875604","wikidata":"https://www.wikidata.org/wiki/Q75","display_name":"The Internet","level":2,"score":0.486868292093277},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4264361262321472},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.330757200717926},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.22296109795570374},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.20436376333236694},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.1717396080493927}],"mesh":[],"locations_count":10,"locations":[{"id":"doi:10.1038/s41586-024-07566-y","is_oa":true,"landing_page_url":"https://doi.org/10.1038/s41586-024-07566-y","pdf_url":"https://www.nature.com/articles/s41586-024-07566-y.pdf","source":{"id":"https://openalex.org/S137773608","display_name":"Nature","issn_l":"0028-0836","issn":["0028-0836","1476-4687"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319908","host_organization_name":"Nature Portfolio","host_organization_lineage":["https://openalex.org/P4310319908","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Nature Portfolio","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Nature","raw_type":"journal-article"},{"id":"pmid:39048682","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/39048682","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Nature","raw_type":null},{"id":"pmh:oai:ora.ox.ac.uk:uuid:fa1155e9-c2ff-436a-8391-455b622f4e64","is_oa":true,"landing_page_url":null,"pdf_url":"https://ora.ox.ac.uk/objects/uuid:fa1155e9-c2ff-436a-8391-455b622f4e64","source":{"id":"https://openalex.org/S4306402636","display_name":"Oxford University Research Archive (ORA) (University of Oxford)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I40120149","host_organization_name":"University of Oxford","host_organization_lineage":["https://openalex.org/I40120149"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Jisc Publications Router","raw_type":"Journal article"},{"id":"pmh:oai:pubmedcentral.nih.gov:11269175","is_oa":true,"landing_page_url":"https://www.ncbi.nlm.nih.gov/pmc/articles/11269175","pdf_url":"https://pmc.ncbi.nlm.nih.gov/articles/PMC11269175/pdf/41586_2024_Article_7566.pdf","source":{"id":"https://openalex.org/S2764455111","display_name":"PubMed Central","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Nature","raw_type":"Text"},{"id":"pmh:oai:www.repository.cam.ac.uk:1810/371811","is_oa":true,"landing_page_url":"https://www.repository.cam.ac.uk/handle/1810/371811","pdf_url":"https://www.repository.cam.ac.uk/bitstreams/5b0bd521-bd16-474f-91b3-263ade59cc55/download","source":{"id":"https://openalex.org/S4306401777","display_name":"Apollo (University of Cambridge)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I241749","host_organization_name":"University of Cambridge","host_organization_lineage":["https://openalex.org/I241749"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"pmh:oai:www.repository.cam.ac.uk:1810/381799","is_oa":true,"landing_page_url":"https://www.repository.cam.ac.uk/handle/1810/381799","pdf_url":"https://www.repository.cam.ac.uk/bitstreams/7621d13d-aa1d-4985-806d-fe34108bff63/download","source":{"id":"https://openalex.org/S4306401777","display_name":"Apollo (University of Cambridge)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I241749","host_organization_name":"University of Cambridge","host_organization_lineage":["https://openalex.org/I241749"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"pmh:oai:www.repository.cam.ac.uk:1810/381841","is_oa":true,"landing_page_url":"https://www.repository.cam.ac.uk/handle/1810/381841","pdf_url":"https://www.repository.cam.ac.uk/bitstreams/130c9f60-1cd0-4f23-9014-dc3870fd90ea/download","source":{"id":"https://openalex.org/S4306401777","display_name":"Apollo (University of Cambridge)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I241749","host_organization_name":"University of Cambridge","host_organization_lineage":["https://openalex.org/I241749"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"pmh:oai:RePEc:nat:nature:v:631:y:2024:i:8022:d:10.1038_s41586-024-07566-y","is_oa":false,"landing_page_url":"https://www.nature.com/articles/s41586-024-07566-y","pdf_url":null,"source":{"id":"https://openalex.org/S4306401271","display_name":"RePEc: Research Papers in Economics","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I77793887","host_organization_name":"Federal Reserve Bank of St. Louis","host_organization_lineage":["https://openalex.org/I77793887"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},{"id":"pmh:oai:pure.ed.ac.uk:openaire/792c0742-46bd-4d3d-b5d4-6a5bdfee2646","is_oa":true,"landing_page_url":"https://www.research.ed.ac.uk/files/460496122/ShumailovEtalNature2024AIModelsCollapseWhen.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306400321","display_name":"Edinburgh Research Explorer (University of Edinburgh)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I98677209","host_organization_name":"University of Edinburgh","host_organization_lineage":["https://openalex.org/I98677209"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Shumailov, I, Shumaylov, Z, Zhao, Y, Papernot, N, Anderson, R & Gal, Y 2024, 'AI models collapse when trained on recursively generated data', Nature, vol. 631, pp. 755-759. https://doi.org/10.1038/s41586-024-07566-y","raw_type":"info:eu-repo/semantics/publishedVersion"},{"id":"pmh:oai:pure.ed.ac.uk:publications/792c0742-46bd-4d3d-b5d4-6a5bdfee2646","is_oa":true,"landing_page_url":"https://www.research.ed.ac.uk/en/publications/792c0742-46bd-4d3d-b5d4-6a5bdfee2646","pdf_url":null,"source":{"id":"https://openalex.org/S4306400321","display_name":"Edinburgh Research Explorer (University of Edinburgh)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I98677209","host_organization_name":"University of Edinburgh","host_organization_lineage":["https://openalex.org/I98677209"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Shumailov, I, Shumaylov, Z, Zhao, Y, Papernot, N, Anderson, R & Gal, Y 2024, 'AI models collapse when trained on recursively generated data', Nature, vol. 631, pp. 755-759. https://doi.org/10.1038/s41586-024-07566-y","raw_type":"info:eu-repo/semantics/publishedVersion"}],"best_oa_location":{"id":"doi:10.1038/s41586-024-07566-y","is_oa":true,"landing_page_url":"https://doi.org/10.1038/s41586-024-07566-y","pdf_url":"https://www.nature.com/articles/s41586-024-07566-y.pdf","source":{"id":"https://openalex.org/S137773608","display_name":"Nature","issn_l":"0028-0836","issn":["0028-0836","1476-4687"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319908","host_organization_name":"Nature Portfolio","host_organization_lineage":["https://openalex.org/P4310319908","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Nature Portfolio","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Nature","raw_type":"journal-article"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.5400000214576721}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4400949264.pdf"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W2090487162","https://openalex.org/W2747329762","https://openalex.org/W2965373594","https://openalex.org/W2982701845","https://openalex.org/W3213748123","https://openalex.org/W4378765257","https://openalex.org/W6778883912","https://openalex.org/W6800751262","https://openalex.org/W6811340617"],"related_works":["https://openalex.org/W2060983174","https://openalex.org/W4365211920","https://openalex.org/W3014948380","https://openalex.org/W4380551139","https://openalex.org/W4317695495","https://openalex.org/W4287117424","https://openalex.org/W4387506531","https://openalex.org/W4238433571","https://openalex.org/W3174044702","https://openalex.org/W2967848559"],"abstract_inverted_index":{")":[0],"demonstrated":[1],"high":[2],"performance":[3],"across":[4],"a":[5],"variety":[6],"of":[7,43,60,70,85,150,162,178],"language":[8,13,30],"tasks.":[9],"ChatGPT":[10],"introduced":[11],"such":[12,27],"models":[14,31,117],"to":[15,35,54,93,146],"the":[16,41,61,79,86,124,148,157,160,176,185],"public.":[17],"It":[18],"is":[19,33],"now":[20],"clear":[21],"that":[22,67,101,137],"generative":[23,133],"artificial":[24],"intelligence":[25],"(AI)":[26],"as":[28,96,107,109],"large":[29],"(LLMs)":[32],"here":[34],"stay":[36],"and":[37,46,99,114,126],"will":[38,171],"substantially":[39],"change":[40],"ecosystem":[42],"online":[44],"text":[45,62],"images.":[47],"Here":[48],"we":[49,144],"consider":[50],"what":[51],"may":[52],"happen":[53],"GPT-{n}":[55],"once":[56],"LLMs":[57,106],"contribute":[58],"much":[59],"found":[63],"online.":[64],"We":[65,91,119,135],"find":[66],"indiscriminate":[68],"use":[69],"model-generated":[71],"content":[72,88,180],"in":[73,78,82,105,110,175,181],"training":[74,151],"causes":[75],"irreversible":[76],"defects":[77],"resulting":[80],"models,":[81],"which":[83],"tails":[84],"original":[87],"distribution":[89],"disappear.":[90],"refer":[92],"this":[94],"effect":[95],"'model":[97],"collapse'":[98],"show":[100],"it":[102,138],"can":[103],"occur":[104],"well":[108],"variational":[111],"autoencoders":[112],"(VAEs)":[113],"Gaussian":[115],"mixture":[116],"(GMMs).":[118],"build":[120],"theoretical":[121],"intuition":[122],"behind":[123],"phenomenon":[125],"portray":[127],"its":[128],"ubiquity":[129],"among":[130],"all":[131],"learned":[132],"models.":[134],"demonstrate":[136],"must":[139],"be":[140,172],"taken":[141],"seriously":[142],"if":[143],"are":[145],"sustain":[147],"benefits":[149],"from":[152,156,184],"large-scale":[153],"data":[154,163,182],"scraped":[155],"web.":[158],"Indeed,":[159],"value":[161],"collected":[164],"about":[165],"genuine":[166],"human":[167],"interactions":[168],"with":[169],"systems":[170],"increasingly":[173],"valuable":[174],"presence":[177],"LLM-generated":[179],"crawled":[183],"Internet.":[186]},"counts_by_year":[{"year":2026,"cited_by_count":177},{"year":2025,"cited_by_count":326},{"year":2024,"cited_by_count":57}],"updated_date":"2026-06-04T09:04:59.091469","created_date":"2025-10-10T00:00:00"}