{"id":"https://openalex.org/W7123353892","doi":"https://doi.org/10.1109/cdc57313.2025.11312306","title":"Preventing Model Collapse when Training LLMs with Synthetic Data","display_name":"Preventing Model Collapse when Training LLMs with Synthetic Data","publication_year":2025,"publication_date":"2025-12-09","ids":{"openalex":"https://openalex.org/W7123353892","doi":"https://doi.org/10.1109/cdc57313.2025.11312306"},"language":null,"primary_location":{"id":"doi:10.1109/cdc57313.2025.11312306","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cdc57313.2025.11312306","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 64th Conference on Decision and Control (CDC)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5022064746","display_name":"Bahman Gharesifard","orcid":"https://orcid.org/0000-0003-1495-3489"},"institutions":[{"id":"https://openalex.org/I204722609","display_name":"Queen's University","ror":"https://ror.org/02y72wh86","country_code":"CA","type":"education","lineage":["https://openalex.org/I204722609"]}],"countries":["CA"],"is_corresponding":false,"raw_author_name":"Bahman Gharesifard","raw_affiliation_strings":["Queen’s University,Department of Mathematics and Statistics,Kingston,ON,Canada"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Queen’s University,Department of Mathematics and Statistics,Kingston,ON,Canada","institution_ids":["https://openalex.org/I204722609"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5114367731","display_name":"Paulo Tabuada","orcid":null},"institutions":[{"id":"https://openalex.org/I161318765","display_name":"University of California, Los Angeles","ror":"https://ror.org/046rm7j60","country_code":"US","type":"education","lineage":["https://openalex.org/I161318765"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Paulo Tabuada","raw_affiliation_strings":["University of California at Los Angeles,Electrical and Computer Engineering Department,Los Angeles,CA,USA,90095"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California at Los Angeles,Electrical and Computer Engineering Department,Los Angeles,CA,USA,90095","institution_ids":["https://openalex.org/I161318765"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.56308965,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1124","last_page":"1129"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.257999986410141,"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"}},"topics":[{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.257999986410141,"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/T11206","display_name":"Model Reduction and Neural Networks","score":0.1526000052690506,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12072","display_name":"Machine Learning and Algorithms","score":0.05920000001788139,"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/synthetic-data","display_name":"Synthetic data","score":0.8216999769210815},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.6201000213623047},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.57669997215271},{"id":"https://openalex.org/keywords/simple","display_name":"Simple (philosophy)","score":0.48159998655319214},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.46799999475479126},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.4675999879837036}],"concepts":[{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.8216999769210815},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.6201000213623047},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.57669997215271},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.49570000171661377},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.48159998655319214},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.46799999475479126},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.4675999879837036},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.41760000586509705},{"id":"https://openalex.org/C24756922","wikidata":"https://www.wikidata.org/wiki/Q1757694","display_name":"Data quality","level":3,"score":0.4120999872684479},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.400299996137619},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.3594000041484833},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33739998936653137},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3228999972343445},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.31369999051094055},{"id":"https://openalex.org/C93226319","wikidata":"https://www.wikidata.org/wiki/Q193137","display_name":"Differential (mechanical device)","level":2,"score":0.30959999561309814},{"id":"https://openalex.org/C55037315","wikidata":"https://www.wikidata.org/wiki/Q5421151","display_name":"Experimental data","level":2,"score":0.3095000088214874},{"id":"https://openalex.org/C100463513","wikidata":"https://www.wikidata.org/wiki/Q5227322","display_name":"Data model (GIS)","level":2,"score":0.3057999908924103},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.3001999855041504}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cdc57313.2025.11312306","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cdc57313.2025.11312306","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 64th Conference on Decision and Control (CDC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"use":[1],"of":[2,8,19,22,81,101,110],"synthetic":[3,36,61,102],"data":[4,37,62,65,105],"in":[5,31],"the":[6,17,32,44,73,79,99,108],"training":[7,25],"generative":[9,83],"models":[10],"is":[11],"a":[12,55,82,92],"common":[13],"approach":[14],"to":[15,40,59,68,106],"address":[16],"lack":[18],"large":[20],"amounts":[21],"high":[23],"quality":[24],"data.":[26],"It":[27],"has":[28],"been":[29],"reported":[30],"literature":[33],"that":[34,78],"using":[35],"can":[38,88],"lead":[39],"model":[41,45,70],"degeneration,":[42],"i.e.,":[43],"stops":[46],"producing":[47],"useful":[48],"outputs.":[49],"In":[50],"this":[51],"paper":[52],"we":[53,76],"propose":[54],"very":[56],"simple":[57],"strategy":[58],"combine":[60],"and":[63,95,103],"human-generated":[64,104],"so":[66],"as":[67],"prevent":[69],"degeneration.":[71],"At":[72],"technical":[74],"level":[75],"show":[77],"evolution":[80],"model,":[84],"caused":[85],"by":[86,91],"re-training,":[87],"be":[89],"described":[90],"differential":[93],"equation":[94],"provide":[96],"conditions":[97],"on":[98],"ratio":[100],"ensure":[107],"existence":[109],"an":[111],"asymptotically":[112],"stable":[113],"equilibrium.":[114]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-01-14T00:00:00"}