{"id":"https://openalex.org/W7167581583","doi":"https://doi.org/10.48550/arxiv.2607.04270","title":"LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation","display_name":"LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation","publication_year":2026,"publication_date":"2026-07-05","ids":{"openalex":"https://openalex.org/W7167581583","doi":"https://doi.org/10.48550/arxiv.2607.04270"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.04270","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.04270","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,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2607.04270","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140196222","display_name":"Hongchen Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Hongchen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140218336","display_name":"Bohao Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Bohao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140164402","display_name":"Jingbang Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Jingbang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140197588","display_name":"Weiqin Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Weiqin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140207087","display_name":"Hang Pan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan, Hang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013997098","display_name":"Bingde Hu","orcid":"https://orcid.org/0000-0003-2556-9239"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Bingde","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140221584","display_name":"Can Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Can","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5140192350","display_name":"Jiawei Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Jiawei","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":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10203","display_name":"Recommender Systems and Techniques","score":0.5795000195503235,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10203","display_name":"Recommender Systems and Techniques","score":0.5795000195503235,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.11550000309944153,"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/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.04100000113248825,"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/security-token","display_name":"Security token","score":0.5846999883651733},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5627999901771545},{"id":"https://openalex.org/keywords/offset","display_name":"Offset (computer science)","score":0.48899999260902405},{"id":"https://openalex.org/keywords/prefix","display_name":"Prefix","score":0.4805000126361847},{"id":"https://openalex.org/keywords/crowdsourcing","display_name":"Crowdsourcing","score":0.45809999108314514},{"id":"https://openalex.org/keywords/recommender-system","display_name":"Recommender system","score":0.44679999351501465},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4120999872684479},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.39649999141693115},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.39320001006126404}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7856000065803528},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5846999883651733},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5627999901771545},{"id":"https://openalex.org/C175291020","wikidata":"https://www.wikidata.org/wiki/Q1156822","display_name":"Offset (computer science)","level":2,"score":0.48899999260902405},{"id":"https://openalex.org/C141603448","wikidata":"https://www.wikidata.org/wiki/Q134830","display_name":"Prefix","level":2,"score":0.4805000126361847},{"id":"https://openalex.org/C62230096","wikidata":"https://www.wikidata.org/wiki/Q275969","display_name":"Crowdsourcing","level":2,"score":0.45809999108314514},{"id":"https://openalex.org/C557471498","wikidata":"https://www.wikidata.org/wiki/Q554950","display_name":"Recommender system","level":2,"score":0.44679999351501465},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.415800005197525},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4120999872684479},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.40369999408721924},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.39649999141693115},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.39320001006126404},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.375900000333786},{"id":"https://openalex.org/C81669768","wikidata":"https://www.wikidata.org/wiki/Q2359161","display_name":"Precision and recall","level":2,"score":0.3666999936103821},{"id":"https://openalex.org/C136886441","wikidata":"https://www.wikidata.org/wiki/Q926129","display_name":"Normalization (sociology)","level":2,"score":0.35499998927116394},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3528999984264374},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.33410000801086426},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.31709998846054077},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.31380000710487366},{"id":"https://openalex.org/C2779458634","wikidata":"https://www.wikidata.org/wiki/Q24963715","display_name":"Debiasing","level":2,"score":0.3037000000476837},{"id":"https://openalex.org/C2776187449","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Natural language generation","level":3,"score":0.3009999990463257},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.2827000021934509},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.2757999897003174},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.27140000462532043},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.25110000371932983},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.250900000333786}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.04270","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.04270","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,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2607.04270","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.04270","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,"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"},"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":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"have":[4],"recently":[5],"emerged":[6],"as":[7,16],"powerful":[8],"backbones":[9],"for":[10,122],"recommender":[11],"systems":[12],"by":[13,37],"reformulating":[14],"recommendation":[15,105,199],"a":[17,26,117,139],"token-level":[18],"generation":[19],"task.":[20],"Despite":[21],"their":[22],"promise,":[23],"we":[24,111],"identify":[25],"pervasive":[27],"yet":[28],"underexplored":[29],"issue:":[30],"$\\textit{Length":[31],"Bias}$.":[32],"Because":[33],"items":[34],"are":[35],"represented":[36],"textual":[38],"descriptions":[39,58],"of":[40,173,215],"varying":[41],"lengths,":[42],"LLM-based":[43,127,187],"recommenders":[44,188],"can":[45,97],"be":[46],"systematically":[47],"biased":[48],"in":[49,62,126],"two":[50,185],"ways.":[51],"On":[52,77],"the":[53,63,78,150,170,174],"input":[54],"side,":[55,80,152],"longer":[56],"item":[57],"occupy":[59],"more":[60],"tokens":[61],"context":[64],"and":[65,102,119,184,201,207],"thus":[66],"receive":[67],"disproportionately":[68],"large":[69],"aggregate":[70],"attention":[71,143,147],"mass":[72],"during":[73],"user":[74],"preference":[75],"modeling.":[76],"output":[79,151],"decoding":[81],"based":[82],"on":[83,179],"summed":[84],"autoregressive":[85],"log-likelihood":[86],"score":[87],"inherently":[88],"disfavors":[89],"long":[90],"items.":[91],"Worse":[92],"still,":[93],"conventional":[94],"length":[95,124,166,194],"normalization":[96],"introduce":[98],"an":[99,164,211],"additional":[100,205],"bias":[101,125,132,195],"even":[103],"degrade":[104],"performance.":[106],"To":[107],"address":[108],"this":[109],"problem,":[110],"propose":[112],"$\\textbf{LBR}$":[113],"($\\textbf{L}$ength":[114],"$\\textbf{B}$ias":[115],"$\\textbf{R}$eduction),":[116],"lightweight":[118],"model-agnostic":[120],"framework":[121],"mitigating":[123],"recommendation.":[128],"LBR":[129,153,191],"mitigates":[130],"input-side":[131],"via":[133],"Length-Aware":[134],"Attention":[135],"Calibration,":[136],"which":[137],"incorporates":[138],"length-dependent":[140],"offset":[141],"into":[142],"logits":[144],"to":[145],"neutralize":[146],"skew.":[148],"For":[149],"introduces":[154],"Effective":[155],"Information":[156],"Length":[157],"Normalization,":[158],"replacing":[159],"naive":[160],"token":[161],"count":[162],"with":[163,203],"information-theoretic":[165],"surrogate":[167],"derived":[168],"from":[169],"branching":[171],"structure":[172],"prefix":[175],"tree.":[176],"Extensive":[177],"experiments":[178],"three":[180],"real-world":[181],"Amazon":[182],"datasets":[183],"representative":[186],"demonstrate":[189],"that":[190],"substantially":[192],"alleviates":[193],"while":[196],"consistently":[197],"improving":[198],"accuracy":[200],"fairness,":[202],"negligible":[204],"training":[206],"inference":[208],"overhead":[209],"(with":[210],"average":[212],"NDCG@5":[213],"gain":[214],"16.82%).":[216],"The":[217],"code":[218],"is":[219],"available":[220],"at":[221],"https://github.com/Void-JackLee/LBR.":[222]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-07-08T00:00:00"}