{"id":"https://openalex.org/W4403577836","doi":"https://doi.org/10.1145/3627673.3680052","title":"Boosting LLM-based Relevance Modeling with Distribution-Aware Robust Learning","display_name":"Boosting LLM-based Relevance Modeling with Distribution-Aware Robust Learning","publication_year":2024,"publication_date":"2024-10-20","ids":{"openalex":"https://openalex.org/W4403577836","doi":"https://doi.org/10.1145/3627673.3680052"},"language":"en","primary_location":{"id":"doi:10.1145/3627673.3680052","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3627673.3680052","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Information and Knowledge Management","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2412.12504","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101423423","display_name":"Hong Liu","orcid":"https://orcid.org/0009-0002-2361-5721"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Hong Liu","raw_affiliation_strings":["Ant Group, Hangzhou, China"],"affiliations":[{"raw_affiliation_string":"Ant Group, Hangzhou, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052961116","display_name":"Saisai Gong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Saisai Gong","raw_affiliation_strings":["Ant Group, Hangzhou, China"],"affiliations":[{"raw_affiliation_string":"Ant Group, Hangzhou, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020603270","display_name":"Yixin Ji","orcid":"https://orcid.org/0009-0006-9309-3323"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yixin Ji","raw_affiliation_strings":["Ant Group, Hangzhou, China"],"affiliations":[{"raw_affiliation_string":"Ant Group, Hangzhou, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057019623","display_name":"Kaixin Wu","orcid":"https://orcid.org/0009-0005-1592-8607"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kaixin Wu","raw_affiliation_strings":["Ant Group, Hangzhou, China"],"affiliations":[{"raw_affiliation_string":"Ant Group, Hangzhou, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101472360","display_name":"Jia Xu","orcid":"https://orcid.org/0009-0004-1163-513X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jia Xu","raw_affiliation_strings":["Ant Group, Hangzhou, China"],"affiliations":[{"raw_affiliation_string":"Ant Group, Hangzhou, China","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5053242349","display_name":"Jinjie Gu","orcid":"https://orcid.org/0000-0001-7596-4945"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinjie Gu","raw_affiliation_strings":["Ant Group, Hangzhou, China"],"affiliations":[{"raw_affiliation_string":"Ant Group, Hangzhou, China","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5101423423"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.2775571,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.77602688,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"4718","last_page":"4725"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9889000058174133,"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":0.9889000058174133,"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/T11550","display_name":"Text and Document Classification Technologies","score":0.9847000241279602,"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/T12205","display_name":"Time Series Analysis and Forecasting","score":0.9692999720573425,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/boosting","display_name":"Boosting (machine learning)","score":0.8378486633300781},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.729500949382782},{"id":"https://openalex.org/keywords/relevance","display_name":"Relevance (law)","score":0.6196268200874329},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4284849166870117},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.41207700967788696}],"concepts":[{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.8378486633300781},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.729500949382782},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.6196268200874329},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4284849166870117},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.41207700967788696},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3627673.3680052","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3627673.3680052","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Information and Knowledge Management","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2412.12504","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2412.12504","pdf_url":"https://arxiv.org/pdf/2412.12504","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2412.12504","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2412.12504","pdf_url":"https://arxiv.org/pdf/2412.12504","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4403577836.pdf"},"referenced_works_count":29,"referenced_works":["https://openalex.org/W2186845332","https://openalex.org/W2286300105","https://openalex.org/W2336445533","https://openalex.org/W2750779823","https://openalex.org/W2918008835","https://openalex.org/W2952370363","https://openalex.org/W2970641574","https://openalex.org/W2998702515","https://openalex.org/W3043376663","https://openalex.org/W3094444847","https://openalex.org/W3135053698","https://openalex.org/W3153427360","https://openalex.org/W3153594491","https://openalex.org/W3154670582","https://openalex.org/W3217305727","https://openalex.org/W4252076394","https://openalex.org/W4285247752","https://openalex.org/W4285294723","https://openalex.org/W4384107234","https://openalex.org/W4385572001","https://openalex.org/W4385573363","https://openalex.org/W4385688501","https://openalex.org/W4385984563","https://openalex.org/W4387841511","https://openalex.org/W4387846503","https://openalex.org/W4389523725","https://openalex.org/W4390043316","https://openalex.org/W4400526908","https://openalex.org/W4401043375"],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W3082059448","https://openalex.org/W4313640622","https://openalex.org/W4387369504","https://openalex.org/W3046775127","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W3107602296"],"abstract_inverted_index":{"With":[0],"the":[1,14,30,41,152,167,176,200,223,235],"rapid":[2],"advancement":[3],"of":[4,16,32,98,154,162,169],"pre-trained":[5],"large":[6],"language":[7],"models":[8],"(LLMs),":[9],"recent":[10],"endeavors":[11],"have":[12],"leveraged":[13],"capabilities":[15],"LLMs":[17,34,53],"in":[18,22,107,122,140],"relevance":[19,42,58,99,138,156,171],"modeling,":[20,172],"resulting":[21],"enhanced":[23],"performance.":[24],"This":[25,182],"is":[26,66],"usually":[27],"done":[28],"through":[29,60],"process":[31],"fine-tuning":[33,61,212],"on":[35],"specifically":[36],"annotated":[37],"datasets":[38],"to":[39,89,94,150,189,214,233],"determine":[40],"between":[43,226],"queries":[44],"and":[45,62,92,219],"items.":[46],"However,":[47],"there":[48],"are":[49,54,195],"two":[50],"limitations":[51],"when":[52,116],"naively":[55],"employed":[56],"for":[57,70,137,204],"modeling":[59,139,157],"inference.":[63],"First,":[64],"it":[65,111],"not":[67,196],"inherently":[68],"efficient":[69],"performing":[71],"nuanced":[72],"tasks":[73],"beyond":[74],"simple":[75],"yes":[76],"or":[77],"no":[78],"answers,":[79],"such":[80],"as":[81],"assessing":[82],"search":[83,108],"relevance.":[84,164],"It":[85],"may":[86],"therefore":[87],"tend":[88],"be":[90],"overconfident":[91],"struggle":[93],"distinguish":[95],"fine-grained":[96,160],"degrees":[97,161],"(e.g.,":[100],"strong":[101],"relevance,":[102,104],"weak":[103],"irrelevance)":[105],"used":[106],"engines.":[109],"Second,":[110],"exhibits":[112],"significant":[113],"performance":[114,224],"degradation":[115],"confronted":[117],"with":[118],"data":[119],"distribution":[120],"shift":[121],"real-world":[123],"scenarios.":[124],"In":[125],"this":[126],"paper,":[127],"we":[128,144,173,208],"propose":[129,175],"a":[130,210],"novel":[131],"Distribution-Aware":[132,177],"Robust":[133],"Learning":[134],"framework":[135],"(DaRL)":[136],"Alipay":[141],"Search.":[142],"Specifically,":[143],"design":[145],"an":[146],"effective":[147],"loss":[148],"function":[149],"enhance":[151],"discriminability":[153],"LLM-based":[155,170],"across":[158],"various":[159],"query-item":[163],"To":[165],"improve":[166,216],"generalizability":[168],"first":[174],"Sample":[178],"Augmentation":[179],"(DASA)":[180],"module.":[181],"module":[183],"utilizes":[184],"out-of-distribution":[185],"(OOD)":[186],"detection":[187],"techniques":[188],"actively":[190],"select":[191],"appropriate":[192],"samples":[193],"that":[194],"well":[197],"covered":[198],"by":[199],"original":[201],"training":[202],"set":[203],"model":[205],"fine-tuning.":[206],"Furthermore,":[207],"adopt":[209],"multi-stage":[211],"strategy":[213],"simultaneously":[215],"in-distribution":[217],"(ID)":[218],"OOD":[220],"performance,":[221],"bridging":[222],"gap":[225],"them.":[227],"DaRL":[228],"has":[229],"been":[230],"deployed":[231],"online":[232],"serve":[234],"Alipay's":[236],"insurance":[237],"product":[238],"search...":[239]},"counts_by_year":[{"year":2025,"cited_by_count":2}],"updated_date":"2026-02-09T09:26:11.010843","created_date":"2025-10-10T00:00:00"}