{"id":"https://openalex.org/W4415316359","doi":"https://doi.org/10.18653/v1/2026.acl-long.1595","title":"Aligning Large Language Models via Fully Self-Synthetic Data","display_name":"Aligning Large Language Models via Fully Self-Synthetic Data","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4415316359","doi":"https://doi.org/10.18653/v1/2026.acl-long.1595"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.acl-long.1595","doi":"https://doi.org/10.18653/v1/2026.acl-long.1595","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1595","pdf_url":"https://aclanthology.org/2026.acl-long.1595.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2026.acl-long.1595.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114159087","display_name":"Shangjian Yin","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I103635307","display_name":"University of California, Riverside","ror":"https://ror.org/03nawhv43","country_code":"US","type":"education","lineage":["https://openalex.org/I103635307"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shangjian Yin","raw_affiliation_strings":["University of California , Riverside ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California , Riverside ,","institution_ids":["https://openalex.org/I103635307"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120049692","display_name":"Zhepei Wei","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhepei Wei","raw_affiliation_strings":["University of Virginia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Virginia","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120049693","display_name":"Xinyu Zhu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xinyu Zhu","raw_affiliation_strings":["University of Virginia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Virginia","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120049694","display_name":"Wei-Lin Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wei-Lin Chen","raw_affiliation_strings":["University of Virginia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Virginia","institution_ids":["https://openalex.org/I51556381"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5120049695","display_name":"Yu Meng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I51556381","display_name":"University of Virginia","ror":"https://ror.org/0153tk833","country_code":"US","type":"education","lineage":["https://openalex.org/I51556381"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yu Meng","raw_affiliation_strings":["University of Virginia"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Virginia","institution_ids":["https://openalex.org/I51556381"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.01124449,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"34553","last_page":"34568"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.41530001163482666,"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":0.41530001163482666,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.2223999947309494,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.05889999866485596,"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/large-language-model-alignment","display_name":"large language model alignment","score":0.9620000123977661},{"id":"https://openalex.org/keywords/reinforcement-learning-from-ai-feedback","display_name":"reinforcement learning from AI feedback","score":0.8130000233650208},{"id":"https://openalex.org/keywords/reinforcement-learning-from-human-feedback","display_name":"reinforcement learning from human feedback","score":0.5320000052452087},{"id":"https://openalex.org/keywords/preference-optimization","display_name":"preference optimization","score":0.9269999861717224},{"id":"https://openalex.org/keywords/self-assessment","display_name":"self-assessment","score":0.609000027179718},{"id":"https://openalex.org/keywords/question-answering","display_name":"question answering","score":0.7829999923706055},{"id":"https://openalex.org/keywords/mathematical-reasoning","display_name":"mathematical reasoning","score":0.9390000104904175}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7986000180244446},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6597999930381775},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5648999810218811},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.5234000086784363},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5095999836921692},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.5034000277519226},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.4595000147819519},{"id":"https://openalex.org/C313442","wikidata":"https://www.wikidata.org/wiki/Q778556","display_name":"Persona","level":2,"score":0.38260000944137573},{"id":"https://openalex.org/C133462117","wikidata":"https://www.wikidata.org/wiki/Q4929239","display_name":"Data collection","level":2,"score":0.3716000020503998},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35120001435279846},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3495999872684479},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.33709999918937683},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3310000002384186},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.3084000051021576},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2955000102519989},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2743000090122223},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.2614000141620636}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.1595","doi":"https://doi.org/10.18653/v1/2026.acl-long.1595","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1595","pdf_url":"https://aclanthology.org/2026.acl-long.1595.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2510.06652","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.06652","pdf_url":"https://arxiv.org/pdf/2510.06652","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},{"id":"doi:10.48550/arxiv.2510.06652","doi":"https://doi.org/10.48550/arxiv.2510.06652","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.06652","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.1595","doi":"https://doi.org/10.18653/v1/2026.acl-long.1595","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1595","pdf_url":"https://aclanthology.org/2026.acl-long.1595.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4415316359.pdf","grobid_xml":"https://content.openalex.org/works/W4415316359.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4403345629","https://openalex.org/W4389519518","https://openalex.org/W7127447921","https://openalex.org/W4400025402","https://openalex.org/W4386755626","https://openalex.org/W4393904674","https://openalex.org/W7010917313","https://openalex.org/W4399510443","https://openalex.org/W4391987791","https://openalex.org/W7160927010"],"abstract_inverted_index":{"Traditional":[0],"reinforcement":[1],"learning":[2],"from":[3,20],"human":[4],"feedback":[5],"(RLHF)":[6],"for":[7,62,103,139,147],"large":[8],"language":[9],"models":[10,41,44],"(LLMs)":[11],"relies":[12],"on":[13,116,126],"expensive":[14],"humanannotated":[15],"datasets,":[16],"while":[17,122],"Reinforcement":[18],"Learning":[19],"AI":[21],"Feedback":[22],"(RLAIF)":[23],"also":[24],"incurs":[25],"significant":[26],"costs,":[27],"requiring":[28],"the":[29,80,86,112,145],"collection":[30],"of":[31],"diverse":[32,95],"prompts":[33,70,96],"and":[34,75,93,97,144],"corresponding":[35],"responses,":[36,74,98],"often":[37],"necessitating":[38],"external":[39],"reward":[40],"or":[42],"proprietary":[43],"like":[45,119],"GPT-4":[46],"to":[47,88],"annotate":[48],"preference":[49,104],"pairs.In":[50],"this":[51],"work,":[52],"we":[53],"introduce":[54],"Self-Alignment":[55],"Optimization":[56],"(SAO),":[57],"a":[58,136],"fully":[59],"self-synthetic":[60],"framework":[61],"LLM":[63,87],"alignment,":[64],"where":[65],"all":[66],"training":[67],"data,":[68],"including":[69],"(i.e.,":[71],"user":[72],"queries),":[73],"preferences,":[76],"are":[77,100],"generated":[78],"by":[79],"model":[81],"itself.Specifically,":[82],"SAO":[83,109],"first":[84],"instructs":[85],"engage":[89],"in":[90,141],"persona":[91],"role-play":[92],"generate":[94],"which":[99],"then":[101],"selfevaluated":[102],"optimization.Extensive":[105],"experiments":[106],"demonstrate":[107],"that":[108],"effectively":[110],"enhances":[111],"model's":[113],"chat":[114],"capabilities":[115],"standard":[117],"benchmarks":[118],"AlpacaEval":[120],"2.0,":[121],"maintaining":[123],"strong":[124],"performance":[125],"downstream":[127],"objective":[128],"tasks":[129],"(e.g.,":[130],"questionanswering,":[131],"math":[132],"reasoning).Our":[133],"work":[134],"provides":[135],"practical":[137],"solution":[138],"self-improvement":[140],"aligning":[142],"LLMs,":[143],"code":[146],"reproducing":[148],"our":[149],"results":[150],"is":[151],"available":[152],"at:":[153],"https://github.":[154],"com/SJY8460/SAO.":[155]},"counts_by_year":[],"updated_date":"2026-10-29T09:56:32.521200","created_date":"2025-10-18T00:00:00"}