{"id":"https://openalex.org/W4412945122","doi":"https://doi.org/10.18653/v1/2025.acl-long.819","title":"RetroLLM: Empowering Large Language Models to Retrieve Fine-grained Evidence within Generation","display_name":"RetroLLM: Empowering Large Language Models to Retrieve Fine-grained Evidence within Generation","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945122","doi":"https://doi.org/10.18653/v1/2025.acl-long.819"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.819","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.819","pdf_url":"https://aclanthology.org/2025.acl-long.819.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.819.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100667935","display_name":"Xiaoxi Li","orcid":"https://orcid.org/0009-0003-0708-418X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaoxi Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000165016","display_name":"Jiajie Jin","orcid":"https://orcid.org/0009-0006-4808-1534"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiajie Jin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016388137","display_name":"Yujia Zhou","orcid":"https://orcid.org/0000-0003-0889-2261"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yujia Zhou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055283292","display_name":"Yongkang Wu","orcid":"https://orcid.org/0000-0003-2086-2356"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yongkang Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5115666591","display_name":"Zhonghua Li","orcid":"https://orcid.org/0000-0002-6378-045X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhonghua Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052177334","display_name":"Qi Ye","orcid":"https://orcid.org/0000-0003-0328-5075"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ye Qi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5010558184","display_name":"Zhicheng Dou","orcid":"https://orcid.org/0000-0002-9781-948X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhicheng Dou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":8.7939,"has_fulltext":true,"cited_by_count":5,"citation_normalized_percentile":{"value":0.97457284,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"16754","last_page":"16779"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9283000230789185,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9283000230789185,"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/computer-science","display_name":"Computer science","score":0.7812783122062683},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.43875959515571594},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.33685770630836487}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7812783122062683},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.43875959515571594},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.33685770630836487}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.819","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.819","pdf_url":"https://aclanthology.org/2025.acl-long.819.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.819","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.819","pdf_url":"https://aclanthology.org/2025.acl-long.819.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1899978318","display_name":null,"funder_award_id":"L233008","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G228125052","display_name":null,"funder_award_id":"62272467","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"},{"id":"https://openalex.org/F4320322499","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92"},{"id":"https://openalex.org/F4320322724","display_name":"Ministry of Education, India","ror":"https://ror.org/048xjjh50"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412945122.pdf","grobid_xml":"https://content.openalex.org/works/W4412945122.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"exhibit":[4],"remarkable":[5],"generative":[6],"capabilities":[7],"but":[8,23],"often":[9],"suffer":[10],"from":[11,39,79],"hallucinations.Retrieval-augmented":[12],"generation":[13,66],"(RAG)":[14],"offers":[15],"an":[16],"effective":[17],"solution":[18],"by":[19],"incorporating":[20],"external":[21],"knowledge,":[22],"existing":[24],"methods":[25],"still":[26],"face":[27],"several":[28],"limitations:":[29],"additional":[30],"deployment":[31],"costs":[32],"of":[33,46,49,92,110,131],"separate":[34],"retrievers,":[35],"redundant":[36],"input":[37],"tokens":[38],"retrieved":[40],"text":[41],"chunks,":[42],"and":[43,51,65,120,151],"the":[44,80,90,129],"lack":[45],"joint":[47],"optimization":[48],"retrieval":[50,64],"generation.To":[52],"address":[53],"these":[54],"issues,":[55],"we":[56,96],"propose":[57],"RetroLLM,":[58],"a":[59,68,108,122],"unified":[60],"framework":[61],"that":[62],"integrates":[63],"into":[67],"single,":[69],"auto-regressive":[70],"process,":[71],"enabling":[72],"LLMs":[73],"to":[74,85,106,134],"directly":[75],"generate":[76,103],"fine-grained":[77],"evidence":[78,94,114,136],"corpus":[81],"with":[82],"constrained":[83,93,124],"decoding.Moreover,":[84],"mitigate":[86],"false":[87],"pruning":[88],"in":[89],"process":[91],"generation,":[95,115],"introduce":[97],"(1)":[98],"hierarchical":[99],"FM-Index":[100],"constraints,":[101],"which":[102,127],"corpus-constrained":[104],"clues":[105],"identify":[107],"subset":[109],"relevant":[111],"documents":[112],"before":[113],"reducing":[116],"irrelevant":[117],"decoding":[118,125],"space;":[119],"(2)":[121],"forward-looking":[123],"strategy,":[126],"considers":[128],"relevance":[130],"future":[132],"sequences":[133],"improve":[135],"accuracy.Extensive":[137],"experiments":[138],"on":[139],"five":[140],"open-domain":[141],"QA":[142],"datasets":[143],"demonstrate":[144],"RetroLLM's":[145],"superior":[146],"performance":[147],"across":[148],"both":[149],"in-domain":[150],"out-ofdomain":[152],"tasks.The":[153],"code":[154],"is":[155],"available":[156],"at":[157],"https:":[158],"//github.com/sunnynexus/RetroLLM.":[159]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":2}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}