{"id":"https://openalex.org/W4412888583","doi":"https://doi.org/10.18653/v1/2025.findings-acl.271","title":"On-Policy Self-Alignment with Fine-grained Knowledge Feedback for Hallucination Mitigation","display_name":"On-Policy Self-Alignment with Fine-grained Knowledge Feedback for Hallucination Mitigation","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888583","doi":"https://doi.org/10.18653/v1/2025.findings-acl.271"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.271","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.271","pdf_url":"https://aclanthology.org/2025.findings-acl.271.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.271.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5029394438","display_name":"Xueru Wen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xueru Wen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068680718","display_name":"Jie Lou","orcid":"https://orcid.org/0000-0001-6639-9913"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jie Lou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101690931","display_name":"Xinyu Lu","orcid":"https://orcid.org/0000-0001-8298-6923"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xinyu Lu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034641436","display_name":"Yuqiu Ji","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuqiu Ji","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051575106","display_name":"Xinyan Guan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xinyan Guan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103090910","display_name":"Yaojie Lu","orcid":"https://orcid.org/0000-0002-5842-7715"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yaojie Lu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100586012","display_name":"Hongyu Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hongyu Lin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100969846","display_name":"Ben He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ben He","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100620300","display_name":"Xianpei Han","orcid":"https://orcid.org/0000-0002-1304-6302"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xianpei Han","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103145743","display_name":"Debing Zhang","orcid":"https://orcid.org/0000-0003-4048-0531"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Debing Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5059068224","display_name":"Le Sun","orcid":"https://orcid.org/0000-0001-6465-8678"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Le Sun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":11,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.9909,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.91024092,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"5215","last_page":"5231"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13283","display_name":"Mental Health Research Topics","score":0.9660000205039978,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T13283","display_name":"Mental Health Research Topics","score":0.9660000205039978,"subfield":{"id":"https://openalex.org/subfields/3205","display_name":"Experimental and Cognitive Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7156396508216858},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3338240385055542},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.321874737739563}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7156396508216858},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3338240385055542},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.321874737739563}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.271","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.271","pdf_url":"https://aclanthology.org/2025.findings-acl.271.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.271","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.271","pdf_url":"https://aclanthology.org/2025.findings-acl.271.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/13","score":0.5299999713897705,"display_name":"Climate action"}],"awards":[{"id":"https://openalex.org/G7398645837","display_name":null,"funder_award_id":"62272439","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888583.pdf","grobid_xml":"https://content.openalex.org/works/W4412888583.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2772917594","https://openalex.org/W2036807459","https://openalex.org/W2058170566","https://openalex.org/W2755342338","https://openalex.org/W2166024367","https://openalex.org/W3116076068","https://openalex.org/W2229312674","https://openalex.org/W2951359407","https://openalex.org/W2079911747","https://openalex.org/W1969923398"],"abstract_inverted_index":{"Hallucination":[0],"occurs":[1],"when":[2],"large":[3],"language":[4],"models":[5,29],"exhibit":[6],"behavior":[7,68],"that":[8,56],"deviates":[9],"from":[10],"the":[11,78,109],"boundaries":[12,64],"of":[13],"their":[14,62,95],"knowledge":[15,63,103],"during":[16],"response":[17],"generation.To":[18],"address":[19],"this":[20,39,85],"critical":[21],"issue,":[22],"previous":[23],"learning-based":[24],"methods":[25],"attempt":[26],"to":[27,59,124],"finetune":[28],"but":[30],"are":[31,88,99,112],"limited":[32],"by":[33],"off-policy":[34],"sampling":[35],"and":[36,65,94,97,127,137],"coarse-grained":[37],"feedback.In":[38],"paper,":[40],"we":[41],"present":[42],"R":[43],"einforcement":[44],"L":[45],"earning":[46],"f":[47],"or":[48],"H":[49],"allucination":[50],"(RLFH),":[51],"an":[52],"on-policy":[53],"selfalignment":[54],"approach":[55],"enables":[57,120],"LLMs":[58],"actively":[60],"explore":[61],"self-correct":[66],"generation":[67],"through":[69],"finegrained":[70],"feedback":[71,107],"signals.RLFH":[72],"introduces":[73],"a":[74],"self-assessment":[75],"framework":[76],"where":[77],"policy":[79],"serves":[80],"as":[81],"its":[82],"own":[83],"judge.Through":[84],"framework,":[86],"responses":[87],"automatically":[89],"decomposed":[90],"into":[91,115],"atomic":[92],"facts":[93],"truthfulness":[96],"informativeness":[98],"assessed":[100],"against":[101],"external":[102],"sources.The":[104],"resulting":[105],"fine-grained":[106],"at":[108],"statement":[110],"level":[111],"then":[113],"converted":[114],"token-level":[116],"dense":[117],"reward":[118],"signals.This":[119],"online":[121],"reinforcement":[122],"learning":[123],"achieve":[125],"precise":[126],"timely":[128],"optimization":[129],"without":[130],"human":[131],"intervention.Comprehensive":[132],"evaluations":[133],"on":[134],"HotpotQA,":[135],"SQuADv2,":[136],"Biography":[138],"benchmarks":[139],"validate":[140],"RLFH's":[141],"effectiveness":[142],"in":[143],"hallucination":[144],"mitigation.":[145]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}