{"id":"https://openalex.org/W7224775928","doi":"https://doi.org/10.48550/arxiv.2610.40314","title":"Decompose, Estimate, Select: Coverage-Aware Context Construction for Long-Form RAG","display_name":"Decompose, Estimate, Select: Coverage-Aware Context Construction for Long-Form RAG","publication_year":2026,"publication_date":"2026-10-29","ids":{"openalex":"https://openalex.org/W7224775928","doi":"https://doi.org/10.48550/arxiv.2610.40314"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.40314","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.40314","pdf_url":"https://arxiv.org/pdf/2610.40314","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.40314","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5160194974","display_name":"He Tu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I258800397","display_name":"New York University Shanghai","ror":"https://ror.org/02vpsdb40","country_code":"CN","type":"education","lineage":["https://openalex.org/I258800397","https://openalex.org/I57206974"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tu, He","raw_affiliation_strings":["New York University Shanghai"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"New York University Shanghai","institution_ids":["https://openalex.org/I258800397"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5116209676","display_name":"Aleena Bosco Kurumthottam","orcid":"https://orcid.org/0009-0007-4915-1749","observed_orcids":["https://orcid.org/0009-0007-4915-1749"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kurumthottam, Aleena Bosco","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5074323303","display_name":"Torsten Suel","orcid":"https://orcid.org/0000-0002-8324-980X","observed_orcids":["https://orcid.org/0000-0002-8324-980X"]},"institutions":[{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Suel, Torsten","raw_affiliation_strings":["New York University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"New York University","institution_ids":["https://openalex.org/I57206974"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.8130000233650208,"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.8130000233650208,"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/T10028","display_name":"Topic Modeling","score":0.17919999361038208,"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":0.0010999999940395355,"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/retrieval-augmented-generation","display_name":"retrieval-augmented generation","score":0.9620000123977661},{"id":"https://openalex.org/keywords/query-decomposition","display_name":"query decomposition","score":0.7369999885559082},{"id":"https://openalex.org/keywords/bi-encoder","display_name":"bi-encoder","score":0.6060000061988831},{"id":"https://openalex.org/keywords/cross-encoder","display_name":"cross-encoder","score":0.8199999928474426},{"id":"https://openalex.org/keywords/document-selection","display_name":"document selection","score":0.7070000171661377},{"id":"https://openalex.org/keywords/faithfulness","display_name":"faithfulness","score":0.6610000133514404}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6998999714851379},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.6010000109672546},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5187000036239624},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.4471000134944916},{"id":"https://openalex.org/C17231256","wikidata":"https://www.wikidata.org/wiki/Q5156540","display_name":"Completeness (order theory)","level":2,"score":0.4284000098705292},{"id":"https://openalex.org/C149629883","wikidata":"https://www.wikidata.org/wiki/Q660926","display_name":"Fraction (chemistry)","level":2,"score":0.4049000144004822},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3855000138282776},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.3822000026702881},{"id":"https://openalex.org/C2780589192","wikidata":"https://www.wikidata.org/wiki/Q7285140","display_name":"Raising (metalworking)","level":2,"score":0.36489999294281006},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.36340001225471497},{"id":"https://openalex.org/C51823790","wikidata":"https://www.wikidata.org/wiki/Q504353","display_name":"Greedy algorithm","level":2,"score":0.358599990606308},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.3197000026702881},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3009999990463257},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.2955000102519989},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.29280000925064087},{"id":"https://openalex.org/C161156560","wikidata":"https://www.wikidata.org/wiki/Q1638872","display_name":"Document retrieval","level":2,"score":0.29159998893737793},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2694999873638153}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.40314","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.40314","pdf_url":"https://arxiv.org/pdf/2610.40314","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2610.40314","doi":"https://doi.org/10.48550/arxiv.2610.40314","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.40314","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.40314","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.40314","pdf_url":"https://arxiv.org/pdf/2610.40314","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7224775928.pdf","grobid_xml":"https://content.openalex.org/works/W7224775928.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7216064583","https://openalex.org/W7162774109","https://openalex.org/W7166673342","https://openalex.org/W7134920483","https://openalex.org/W7118575110","https://openalex.org/W7164536448","https://openalex.org/W7167079065","https://openalex.org/W7219912391","https://openalex.org/W7140196186","https://openalex.org/W7134924563"],"abstract_inverted_index":{"Retrieval-augmented":[0],"generation":[1],"(RAG)":[2],"grounds":[3],"a":[4,35,47,51,56,146],"language":[5],"model's":[6],"output":[7],"in":[8,150],"retrieved":[9],"evidence,":[10,81],"but":[11,132],"for":[12],"complex":[13],"questions":[14],"spanning":[15],"multiple":[16],"aspects":[17,31],"of":[18,171],"an":[19],"information":[20],"need,":[21],"relevance-based":[22],"retrieval":[23,37,105],"may":[24],"return":[25],"redundant":[26,80],"evidence":[27,129,156],"while":[28,185],"leaving":[29],"some":[30],"unsupported.":[32],"We":[33,95],"propose":[34],"coverage-aware":[36],"pipeline":[38],"that":[39,98,121],"decomposes":[40],"queries":[41],"into":[42],"subquestions,":[43],"ranks":[44],"candidates":[45],"with":[46,55],"coverage-trained":[48],"bi-encoder,":[49],"predicts":[50],"document\\,$\\times$\\,subquestion":[52],"coverage":[53,111],"matrix":[54],"cross-encoder,":[57],"and":[58,78,161],"selects":[59],"documents":[60],"by":[61,177,180],"greedy":[62],"maximum-marginal":[63],"coverage.":[64],"On":[65],"TREC":[66],"RAG":[67],"2024,":[68],"our":[69,116],"approach":[70],"improves":[71],"$\u03b1$-nDCG@10,":[72],"which":[73,151],"rewards":[74],"newly":[75],"covered":[76],"nuggets":[77,174],"discounts":[79],"from":[82,157],"$0.625$":[83],"to":[84,112],"$0.702$,":[85],"approaching":[86],"the":[87,113,137,152,169,178],"$0.718$":[88],"gold-nugget":[89],"upper":[90],"bound":[91],"without":[92,164],"sacrificing":[93],"relevance.":[94],"then":[96],"find":[97],"ground-truth":[99],"oracle":[100],"retrieval,":[101],"despite":[102],"substantially":[103],"higher":[104],"coverage,":[106],"delivers":[107],"no":[108],"more":[109],"nugget":[110],"answer":[114,122,160],"than":[115],"estimated":[117],"selection.":[118],"This":[119],"suggests":[120],"completeness":[123],"depends":[124],"not":[125],"only":[126],"on":[127,134],"what":[128],"is":[130],"retrieved,":[131],"also":[133],"how":[135],"effectively":[136],"generator":[138],"uses":[139],"it.":[140],"To":[141],"address":[142],"this,":[143],"we":[144],"introduce":[145],"\\emph{grounded":[147],"self-feedback":[148],"loop}":[149],"model":[153],"identifies":[154],"underused":[155],"its":[158],"first":[159],"generates":[162],"again":[163],"gold":[165],"supervision,":[166],"raising":[167],"$V_{\\mathrm{strict}}$,":[168],"fraction":[170],"vital":[172],"reference":[173],"fully":[175],"supported":[176],"answer,":[179],"$0.086$":[181],"over":[182],"two":[183],"passes":[184],"largely":[186],"preserving":[187],"faithfulness.":[188]},"counts_by_year":[],"updated_date":"2026-11-02T09:45:03.842371","created_date":"2026-10-31T00:00:00"}