{"id":"https://openalex.org/W7135229442","doi":"https://doi.org/10.48550/arxiv.2603.11650","title":"QChunker: Learning Question-Aware Text Chunking for Domain RAG via Multi-Agent Debate","display_name":"QChunker: Learning Question-Aware Text Chunking for Domain RAG via Multi-Agent Debate","publication_year":2026,"publication_date":"2026-03-12","ids":{"openalex":"https://openalex.org/W7135229442","doi":"https://doi.org/10.48550/arxiv.2603.11650"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.11650","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.11650","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.11650","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114378701","display_name":"Jihao Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Zhao, Jihao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028037459","display_name":"Daixuan Li","orcid":"https://orcid.org/0009-0008-1996-7939"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Daixuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128947912","display_name":"Pengfei Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Pengfei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129063495","display_name":"Shuaishuai Zu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zu, Shuaishuai","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128958310","display_name":"Biao Qin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qin, Biao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128935042","display_name":"Hongyan Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Hongyan","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5114378701"],"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":0.8500000238418579,"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":0.8500000238418579,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.02250000089406967,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.020400000736117363,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/chunking","display_name":"Chunking (psychology)","score":0.7871999740600586},{"id":"https://openalex.org/keywords/coherence","display_name":"Coherence (philosophical gambling strategy)","score":0.4767000079154968},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.43160000443458557},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.43149998784065247},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.4120999872684479},{"id":"https://openalex.org/keywords/granularity","display_name":"Granularity","score":0.3962000012397766},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.3772999942302704},{"id":"https://openalex.org/keywords/structured-prediction","display_name":"Structured prediction","score":0.352400004863739},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.34310001134872437},{"id":"https://openalex.org/keywords/annotation","display_name":"Annotation","score":0.3368000090122223}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7950999736785889},{"id":"https://openalex.org/C203357204","wikidata":"https://www.wikidata.org/wiki/Q1089605","display_name":"Chunking (psychology)","level":2,"score":0.7871999740600586},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5740000009536743},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5253999829292297},{"id":"https://openalex.org/C2781181686","wikidata":"https://www.wikidata.org/wiki/Q4226068","display_name":"Coherence (philosophical gambling strategy)","level":2,"score":0.4767000079154968},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.43160000443458557},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.43149998784065247},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.4120999872684479},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.3962000012397766},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.3772999942302704},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.352400004863739},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.34310001134872437},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.3368000090122223},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.32659998536109924},{"id":"https://openalex.org/C152565575","wikidata":"https://www.wikidata.org/wiki/Q1124538","display_name":"Conditional random field","level":2,"score":0.3174000084400177},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3111000061035156},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.30820000171661377},{"id":"https://openalex.org/C95318506","wikidata":"https://www.wikidata.org/wiki/Q6588467","display_name":"Textual entailment","level":3,"score":0.30230000615119934},{"id":"https://openalex.org/C2779135771","wikidata":"https://www.wikidata.org/wiki/Q403574","display_name":"Named-entity recognition","level":3,"score":0.29580000042915344},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.29170000553131104},{"id":"https://openalex.org/C2985684807","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Text generation","level":2,"score":0.27889999747276306},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2782999873161316},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2752000093460083},{"id":"https://openalex.org/C98501671","wikidata":"https://www.wikidata.org/wiki/Q1948408","display_name":"Text segmentation","level":3,"score":0.27459999918937683},{"id":"https://openalex.org/C134752490","wikidata":"https://www.wikidata.org/wiki/Q374182","display_name":"Logical consequence","level":2,"score":0.27059999108314514},{"id":"https://openalex.org/C33762810","wikidata":"https://www.wikidata.org/wiki/Q461671","display_name":"Data integrity","level":2,"score":0.26989999413490295},{"id":"https://openalex.org/C2776434776","wikidata":"https://www.wikidata.org/wiki/Q19246213","display_name":"Domain adaptation","level":3,"score":0.26969999074935913},{"id":"https://openalex.org/C2474386","wikidata":"https://www.wikidata.org/wiki/Q461183","display_name":"Text corpus","level":2,"score":0.2689000070095062},{"id":"https://openalex.org/C155092808","wikidata":"https://www.wikidata.org/wiki/Q182557","display_name":"Computational linguistics","level":2,"score":0.262800008058548},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2619999945163727},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.25369998812675476},{"id":"https://openalex.org/C2776960227","wikidata":"https://www.wikidata.org/wiki/Q2586354","display_name":"Knowledge transfer","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.11650","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.11650","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.11650","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.11650","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities","score":0.6428121328353882}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"effectiveness":[1],"upper":[2],"bound":[3],"of":[4,18,52,65,122,177],"retrieval-augmented":[5],"generation":[6],"(RAG)":[7],"is":[8],"fundamentally":[9],"constrained":[10],"by":[11,219],"the":[12,35,45,60,75,103,175,182,200],"semantic":[13],"integrity":[14,64,94],"and":[15,55,63,96,125,139,164,172,198,226],"information":[16],"granularity":[17],"text":[19,46,53,66,92,178,183,228],"chunks":[20,197],"in":[21,142],"its":[22],"knowledge":[23,56,97],"base.":[24],"To":[25],"address":[26],"these":[27],"challenges,":[28],"this":[29,114,127],"paper":[30],"proposes":[31],"QChunker,":[32],"which":[33,147],"restructures":[34],"RAG":[36,221],"paradigm":[37],"from":[38,70],"retrieval-augmentation":[39],"to":[40,58,129,134,193],"understanding-retrieval-augmentation.":[41],"Firstly,":[42],"QChunker":[43,214],"models":[44],"chunking":[47,144],"as":[48,108],"a":[49,80,88,119,156],"composite":[50],"task":[51],"segmentation":[54,184],"completion":[57],"ensure":[59],"logical":[61],"coherence":[62],"chunks.":[67,179,229],"Drawing":[68],"inspiration":[69],"Hal":[71],"Gregersen's":[72],"\"Questions":[73],"Are":[74],"Answer\"":[76],"theory,":[77],"we":[78,116,154,186],"design":[79],"multi-agent":[81],"debate":[82],"framework":[83,100],"comprising":[84],"four":[85,209],"specialized":[86],"components:":[87],"question":[89],"outline":[90],"generator,":[91],"segmenter,":[93],"reviewer,":[95],"completer.":[98],"This":[99],"operates":[101],"on":[102,150],"principle":[104],"that":[105,168,213],"questions":[106],"serve":[107],"catalysts":[109],"for":[110,190],"profound":[111],"insights.":[112],"Through":[113],"pipeline,":[115],"successfully":[117],"construct":[118],"high-quality":[120],"dataset":[121],"45K":[123],"entries":[124],"transfer":[126],"capability":[128],"small":[130],"language":[131],"models.":[132],"Additionally,":[133],"handle":[135],"long":[136],"evaluation":[137,145,159],"chains":[138],"low":[140],"efficiency":[141],"existing":[143],"methods,":[146],"overly":[148],"rely":[149],"downstream":[151],"QA":[152],"tasks,":[153],"introduce":[155],"novel":[157],"direct":[158],"metric,":[160],"ChunkScore.":[161,204],"Both":[162],"theoretical":[163],"experimental":[165,206],"validations":[166],"demonstrate":[167],"ChunkScore":[169],"can":[170],"directly":[171],"efficiently":[173],"discriminate":[174],"quality":[176],"Furthermore,":[180],"during":[181],"phase,":[185],"utilize":[187],"document":[188],"outlines":[189],"multi-path":[191],"sampling":[192],"generate":[194],"multiple":[195],"candidate":[196],"select":[199],"optimal":[201],"solution":[202],"employing":[203],"Extensive":[205],"results":[207],"across":[208],"heterogeneous":[210],"domains":[211],"exhibit":[212],"effectively":[215],"resolves":[216],"aforementioned":[217],"issues":[218],"providing":[220],"with":[222],"more":[223],"logically":[224],"coherent":[225],"information-rich":[227]},"counts_by_year":[],"updated_date":"2026-03-14T06:46:50.379900","created_date":"2026-03-14T00:00:00"}