{"id":"https://openalex.org/W4412889819","doi":"https://doi.org/10.18653/v1/2025.acl-long.1333","title":"Understanding Impact of Human Feedback via Influence Functions","display_name":"Understanding Impact of Human Feedback via Influence Functions","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412889819","doi":"https://doi.org/10.18653/v1/2025.acl-long.1333"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.1333","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1333","pdf_url":"https://aclanthology.org/2025.acl-long.1333.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":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.1333.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5119181876","display_name":"Taywon Min","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Taywon Min","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119181877","display_name":"Haeone Lee","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Haeone Lee","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119181878","display_name":"Yongchan Kwon","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yongchan Kwon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5119181879","display_name":"Kimin Lee","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kimin Lee","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"27471","last_page":"27500"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11810","display_name":"Complex Systems and Decision Making","score":0.4147999882698059,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T11810","display_name":"Complex Systems and Decision Making","score":0.4147999882698059,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5787825584411621}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5787825584411621}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.1333","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1333","pdf_url":"https://aclanthology.org/2025.acl-long.1333.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.1333","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1333","pdf_url":"https://aclanthology.org/2025.acl-long.1333.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":[],"funders":[{"id":"https://openalex.org/F4320320671","display_name":"National Research Foundation","ror":"https://ror.org/05s0g1g46"},{"id":"https://openalex.org/F4320322120","display_name":"National Research Foundation of Korea","ror":"https://ror.org/013aysd81"},{"id":"https://openalex.org/F4320328359","display_name":"Ministry of Science and ICT, South Korea","ror":"https://ror.org/01wpjm123"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412889819.pdf","grobid_xml":"https://content.openalex.org/works/W4412889819.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":{"In":[0],"Reinforcement":[1],"Learning":[2],"from":[3,15],"Human":[4],"Feedback":[5],"(RLHF),":[6],"it":[7],"is":[8],"crucial":[9],"to":[10,18,44,67,92,124,147],"learn":[11],"suitable":[12],"reward":[13,46,78,94],"models":[14,22,95],"human":[16,25,27,72,113,134],"feedback":[17,28,41,73,114,143],"align":[19,126],"large":[20],"language":[21],"(LLMs)":[23],"with":[24,127],"intentions.However,":[26],"can":[29,42,141],"often":[30],"be":[31],"noisy,":[32],"inconsistent,":[33],"or":[34],"biased,":[35],"especially":[36],"when":[37],"evaluating":[38],"complex":[39],"responses.Such":[40],"lead":[43],"misaligned":[45],"signals,":[47],"potentially":[48],"causing":[49],"unintended":[50],"side":[51],"effects":[52],"during":[53],"the":[54,62,69,75,87,131],"RLHF":[55],"process.To":[56],"address":[57],"these":[58],"challenges,":[59],"we":[60,136],"explore":[61],"use":[63],"of":[64,71,77,89,105,133],"influence":[65,90,106,139],"functions":[66,91,140],"measure":[68],"impact":[70,132],"on":[74],"performance":[76],"models.We":[79],"propose":[80],"a":[81],"compute-efficient":[82],"approximation":[83],"method":[84],"that":[85,138],"enables":[86],"application":[88],"LLM-based":[93],"and":[96,116,145,157],"large-scale":[97],"preference":[98],"datasets.Our":[99],"experiments":[100],"showcase":[101],"two":[102],"key":[103],"applications":[104],"functions:(1)":[107],"detecting":[108],"common":[109],"labeler":[110],"biases":[111],"in":[112,120,150],"datasets":[115],"(2)":[117],"guiding":[118],"labelers":[119,153],"refining":[121],"their":[122],"strategies":[123],"better":[125],"expert":[128],"feedback.By":[129],"quantifying":[130],"feedback,":[135],"believe":[137],"enhance":[142],"interpretability":[144],"contribute":[146],"scalable":[148],"oversight":[149],"RLHF,":[151],"helping":[152],"provide":[154],"more":[155],"accurate":[156],"consistent":[158],"feedback.":[159]},"counts_by_year":[{"year":2026,"cited_by_count":3}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}