{"id":"https://openalex.org/W7131883812","doi":"https://doi.org/10.48550/arxiv.2602.22225","title":"SmartChunk Retrieval: Query-Aware Chunk Compression with Planning for Efficient Document RAG","display_name":"SmartChunk Retrieval: Query-Aware Chunk Compression with Planning for Efficient Document RAG","publication_year":2025,"publication_date":"2025-12-17","ids":{"openalex":"https://openalex.org/W7131883812","doi":"https://doi.org/10.48550/arxiv.2602.22225"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2602.22225","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.22225","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.2602.22225","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5127282755","display_name":"Xuechen Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Zhang, Xuechen","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125704338","display_name":"Koustava Goswami","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Goswami, Koustava","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126180976","display_name":"Samet Oymak","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Oymak, Samet","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061832197","display_name":"Jiasi Chen","orcid":"https://orcid.org/0000-0001-9923-9027"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Jiasi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5016933602","display_name":"Nedim Lipka","orcid":"https://orcid.org/0000-0002-3779-7784"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lipka, Nedim","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5127282755"],"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":0.48730000853538513,"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":0.48730000853538513,"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":0.19359999895095825,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.11349999904632568,"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/scalability","display_name":"Scalability","score":0.6561999917030334},{"id":"https://openalex.org/keywords/chunking","display_name":"Chunking (psychology)","score":0.4884999990463257},{"id":"https://openalex.org/keywords/abstraction","display_name":"Abstraction","score":0.4627000093460083},{"id":"https://openalex.org/keywords/granularity","display_name":"Granularity","score":0.45660001039505005},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.45320001244544983},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.4260999858379364},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4163999855518341},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.40130001306533813},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.3944999873638153}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8855999708175659},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6561999917030334},{"id":"https://openalex.org/C203357204","wikidata":"https://www.wikidata.org/wiki/Q1089605","display_name":"Chunking (psychology)","level":2,"score":0.4884999990463257},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.4627000093460083},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46059998869895935},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.45660001039505005},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.45320001244544983},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4260999858379364},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4163999855518341},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.40130001306533813},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.3944999873638153},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.37950000166893005},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3765999972820282},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3698999881744385},{"id":"https://openalex.org/C2776999362","wikidata":"https://www.wikidata.org/wiki/Q2349274","display_name":"Planner","level":2,"score":0.3637000024318695},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3537999987602234},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.35030001401901245},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.34439998865127563},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.33329999446868896},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.32510000467300415},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.3190000057220459},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.31859999895095825},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.31209999322891235},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.3098999857902527},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2994999885559082},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.2840999960899353},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.2648000121116638},{"id":"https://openalex.org/C2474386","wikidata":"https://www.wikidata.org/wiki/Q461183","display_name":"Text corpus","level":2,"score":0.26429998874664307},{"id":"https://openalex.org/C161156560","wikidata":"https://www.wikidata.org/wiki/Q1638872","display_name":"Document retrieval","level":2,"score":0.25450000166893005},{"id":"https://openalex.org/C35639132","wikidata":"https://www.wikidata.org/wiki/Q7452468","display_name":"Sequence labeling","level":3,"score":0.25119999051094055}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2602.22225","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.22225","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.2602.22225","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.22225","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Retrieval-augmented":[0],"generation":[1,48],"(RAG)":[2],"has":[3],"strong":[4,203],"potential":[5],"for":[6,89,109,221],"producing":[7],"accurate":[8],"and":[9,32,47,76,91,112,138,163,175,208],"factual":[10],"outputs":[11],"by":[12,29],"combining":[13],"language":[14],"models":[15],"(LMs)":[16],"with":[17,136,205],"evidence":[18],"retrieved":[19,45],"from":[20,71],"large":[21,80],"text":[22],"corpora.":[23,81],"However,":[24],"current":[25],"pipelines":[26],"are":[27,36,44,53],"limited":[28],"static":[30],"chunking":[31],"flat":[33],"retrieval:":[34],"documents":[35],"split":[37],"into":[38],"short,":[39],"predetermined,":[40],"fixed-size":[41],"chunks,":[42,75],"embeddings":[43,122],"uniformly,":[46],"relies":[49],"on":[50,130,181,211],"whatever":[51],"chunks":[52],"returned.":[54],"This":[55],"design":[56],"brings":[57],"challenges,":[58],"as":[59,217],"retrieval":[60,128],"quality":[61],"is":[62],"highly":[63],"sensitive":[64],"to":[65,79],"chunk":[66,106,121,151],"size,":[67],"often":[68],"introduces":[69],"noise":[70],"irrelevant":[72],"or":[73],"misleading":[74],"scales":[77],"poorly":[78],"We":[82],"present":[83],"SmartChunk":[84,97,133,180,192],"retrieval,":[85],"a":[86,100,114,154,218],"query-adaptive":[87],"framework":[88,220],"efficient":[90],"robust":[92],"long-document":[93],"question":[94],"answering":[95],"(QA).":[96],"uses":[98],"(i)":[99],"planner":[101,147],"that":[102,118],"predicts":[103],"the":[104,131,140],"optimal":[105],"abstraction":[107],"level":[108],"each":[110],"query,":[111],"(ii)":[113],"lightweight":[115],"compression":[116],"module":[117],"produces":[119],"high-level":[120],"without":[123],"repeated":[124],"summarization.":[125],"By":[126],"adapting":[127],"granularity":[129],"fly,":[132],"balances":[134],"accuracy":[135,162],"efficiency":[137],"avoids":[139],"drawbacks":[141],"of":[142],"fixed":[143],"strategies.":[144],"Notably,":[145],"our":[146],"can":[148],"reason":[149],"about":[150],"abstractions":[152],"through":[153],"novel":[155],"reinforcement":[156],"learning":[157],"scheme,":[158],"STITCH,":[159],"which":[160],"boosts":[161],"generalization.":[164],"To":[165],"reflect":[166],"real-world":[167],"applications,":[168],"where":[169],"users":[170],"face":[171],"diverse":[172],"document":[173],"types":[174],"query":[176],"styles,":[177],"we":[178],"evaluate":[179],"five":[182],"QA":[183],"benchmarks":[184],"plus":[185],"one":[186],"out-of-domain":[187,212],"dataset.":[188],"Across":[189],"these":[190],"evaluations,":[191],"outperforms":[193],"state-of-the-art":[194],"RAG":[195],"baselines,":[196],"while":[197],"reducing":[198],"cost.":[199],"Further":[200],"analysis":[201],"demonstrates":[202],"scalability":[204],"larger":[206],"corpora":[207],"consistent":[209],"gains":[210],"datasets,":[213],"highlighting":[214],"its":[215],"effectiveness":[216],"general":[219],"adaptive":[222],"retrieval.":[223]},"counts_by_year":[],"updated_date":"2026-02-28T06:18:59.386488","created_date":"2026-02-28T00:00:00"}