{"id":"https://openalex.org/W7152124379","doi":"https://doi.org/10.48550/arxiv.2604.06163","title":"Data, Not Model: Explaining Bias toward LLM Texts in Neural Retrievers","display_name":"Data, Not Model: Explaining Bias toward LLM Texts in Neural Retrievers","publication_year":2026,"publication_date":"2026-04-07","ids":{"openalex":"https://openalex.org/W7152124379","doi":"https://doi.org/10.48550/arxiv.2604.06163"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.06163","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.06163","pdf_url":null,"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","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.2604.06163","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133200122","display_name":"Wei Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133204302","display_name":"Keping Bi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bi, Keping","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084642830","display_name":"Yinqiong Cai","orcid":"https://orcid.org/0000-0002-7869-8213"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cai, Yinqiong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133161622","display_name":"Wei Chen","orcid":"https://orcid.org/0009-0007-4543-7344"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109736354","display_name":"Jiafeng Guo","orcid":"https://orcid.org/0000-0002-9509-8674"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Jiafeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133176731","display_name":"Xueqi Cheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cheng, Xueqi","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/T10028","display_name":"Topic Modeling","score":0.21549999713897705,"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":0.21549999713897705,"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.16300000250339508,"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/T10918","display_name":"Memory Processes and Influences","score":0.16249999403953552,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/mirroring","display_name":"Mirroring","score":0.717199981212616},{"id":"https://openalex.org/keywords/artifact","display_name":"Artifact (error)","score":0.7006000280380249},{"id":"https://openalex.org/keywords/fluency","display_name":"Fluency","score":0.6431000232696533},{"id":"https://openalex.org/keywords/perspective","display_name":"Perspective (graphical)","score":0.569100022315979},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.4462999999523163},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.43389999866485596},{"id":"https://openalex.org/keywords/false-positive-paradox","display_name":"False positive paradox","score":0.4275999963283539},{"id":"https://openalex.org/keywords/raising","display_name":"Raising (metalworking)","score":0.40299999713897705}],"concepts":[{"id":"https://openalex.org/C189645446","wikidata":"https://www.wikidata.org/wiki/Q350865","display_name":"Mirroring","level":2,"score":0.717199981212616},{"id":"https://openalex.org/C2779010991","wikidata":"https://www.wikidata.org/wiki/Q2720909","display_name":"Artifact (error)","level":2,"score":0.7006000280380249},{"id":"https://openalex.org/C2777413886","wikidata":"https://www.wikidata.org/wiki/Q3276013","display_name":"Fluency","level":2,"score":0.6431000232696533},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.569100022315979},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5498999953269958},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.5078999996185303},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.49459999799728394},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.4462999999523163},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4462999999523163},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.43389999866485596},{"id":"https://openalex.org/C64869954","wikidata":"https://www.wikidata.org/wiki/Q1859747","display_name":"False positive paradox","level":2,"score":0.4275999963283539},{"id":"https://openalex.org/C2780589192","wikidata":"https://www.wikidata.org/wiki/Q7285140","display_name":"Raising (metalworking)","level":2,"score":0.40299999713897705},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.39890000224113464},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.3716000020503998},{"id":"https://openalex.org/C164614171","wikidata":"https://www.wikidata.org/wiki/Q5204775","display_name":"DECIPHER","level":2,"score":0.37119999527931213},{"id":"https://openalex.org/C148609458","wikidata":"https://www.wikidata.org/wiki/Q7021281","display_name":"Nexus (standard)","level":2,"score":0.3474000096321106},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.31929999589920044},{"id":"https://openalex.org/C57493831","wikidata":"https://www.wikidata.org/wiki/Q3134666","display_name":"Projection (relational algebra)","level":2,"score":0.3165999948978424},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.31310001015663147},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3124000132083893},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.29440000653266907},{"id":"https://openalex.org/C75917345","wikidata":"https://www.wikidata.org/wiki/Q2725298","display_name":"Sampling bias","level":3,"score":0.2720000147819519},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2702000141143799},{"id":"https://openalex.org/C79585631","wikidata":"https://www.wikidata.org/wiki/Q431498","display_name":"Confirmation bias","level":2,"score":0.2630999982357025},{"id":"https://openalex.org/C2778095710","wikidata":"https://www.wikidata.org/wiki/Q6031225","display_name":"Information source (mathematics)","level":2,"score":0.2583000063896179},{"id":"https://openalex.org/C2779458634","wikidata":"https://www.wikidata.org/wiki/Q24963715","display_name":"Debiasing","level":2,"score":0.25110000371932983}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.06163","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.06163","pdf_url":null,"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","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.2604.06163","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.06163","pdf_url":null,"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","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":[{"score":0.8060638308525085,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"studies":[1],"show":[2,112],"that":[3,53,69,113],"neural":[4],"retrievers":[5,114],"often":[6],"display":[7],"source":[8,54,168],"bias,":[9],"favoring":[10],"passages":[11],"generated":[12],"by":[13,51,156],"LLMs":[14],"over":[15,132],"human-written":[16,106],"ones,":[17],"even":[18],"when":[19],"both":[20],"are":[21],"semantically":[22],"similar.":[23],"This":[24],"bias":[25,55,95,162],"has":[26],"been":[27],"considered":[28],"an":[29],"inherent":[30],"flaw":[31],"of":[32,41],"retrievers,":[33],"raising":[34],"concerns":[35,176],"about":[36],"the":[37,64,91,94,103,117,121,137,161],"fairness":[38],"and":[39,74,80,87,150],"reliability":[40],"modern":[42],"information":[43,181],"access":[44,182],"systems.":[45,183],"Our":[46],"work":[47],"challenges":[48],"this":[49],"view":[50],"showing":[52],"stems":[56],"from":[57,97,105],"supervision":[58],"in":[59,120,147,180],"retrieval":[60],"datasets":[61],"rather":[62],"than":[63],"models":[65],"themselves.":[66],"We":[67,110,170],"found":[68],"non-semantic":[70],"differences,":[71],"like":[72],"fluency":[73],"term":[75],"specificity,":[76],"exist":[77],"between":[78,85],"positive":[79],"negative":[81],"documents,":[82],"mirroring":[83],"differences":[84,146],"LLM":[86,133,153],"human":[88],"texts.":[89,109,134],"In":[90],"embedding":[92],"space,":[93],"direction":[96,104],"negatives":[98],"to":[99,107,129],"positives":[100],"aligns":[101],"with":[102],"LLM-generated":[108,178],"theoretically":[111],"inevitably":[115],"absorb":[116],"artifact":[118,145],"imbalances":[119],"training":[122,148],"data":[123,149],"during":[124],"contrastive":[125],"learning,":[126],"which":[127],"leads":[128],"their":[130,158],"preferences":[131],"To":[135],"mitigate":[136],"effect,":[138],"we":[139],"propose":[140],"two":[141],"approaches:":[142],"1)":[143],"reducing":[144],"2)":[151],"adjusting":[152],"text":[154],"vectors":[155],"removing":[157],"projection":[159],"on":[160],"vector.":[163],"Both":[164],"methods":[165],"substantially":[166],"reduce":[167],"bias.":[169],"hope":[171],"our":[172],"study":[173],"alleviates":[174],"some":[175],"regarding":[177],"texts":[179]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-09T00:00:00"}