{"id":"https://openalex.org/W7164848537","doi":"https://doi.org/10.1145/3816713.3819507","title":"Hallucination in Retrieval-Augmented Generation is a Context Construction Problem: A Constraint-Driven Cross-Modal Framework","display_name":"Hallucination in Retrieval-Augmented Generation is a Context Construction Problem: A Constraint-Driven Cross-Modal Framework","publication_year":2026,"publication_date":"2026-06-15","ids":{"openalex":"https://openalex.org/W7164848537","doi":"https://doi.org/10.1145/3816713.3819507"},"language":null,"primary_location":{"id":"doi:10.1145/3816713.3819507","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3816713.3819507","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 14th International Conference on Advances in Information Technology","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3816713.3819507","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5021212806","display_name":"Yinxing Li","orcid":"https://orcid.org/0000-0001-9335-9802"},"institutions":[{"id":"https://openalex.org/I201537933","display_name":"Tohoku University","ror":"https://ror.org/01dq60k83","country_code":"JP","type":"education","lineage":["https://openalex.org/I201537933"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Yinxing Li","raw_affiliation_strings":["Graduate School of Economics and Management, Tohoku University, Sendai, Miyagi, Japan"],"raw_orcid":"https://orcid.org/0000-0001-9335-9802","affiliations":[{"raw_affiliation_string":"Graduate School of Economics and Management, Tohoku University, Sendai, Miyagi, Japan","institution_ids":["https://openalex.org/I201537933"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5053294807","display_name":"Nobuhiko Terui","orcid":"https://orcid.org/0000-0003-4868-0140"},"institutions":[{"id":"https://openalex.org/I201537933","display_name":"Tohoku University","ror":"https://ror.org/01dq60k83","country_code":"JP","type":"education","lineage":["https://openalex.org/I201537933"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Victor Okolie","raw_affiliation_strings":["Graduate School of Economics and Management, Tohoku University, Sendai, Miyagi, Japan"],"raw_orcid":"https://orcid.org/0000-0003-4868-0140","affiliations":[{"raw_affiliation_string":"Graduate School of Economics and Management, Tohoku University, Sendai, Miyagi, Japan","institution_ids":["https://openalex.org/I201537933"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I201537933"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.89775631,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.5601999759674072,"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":0.5601999759674072,"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":0.2535000145435333,"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"}},{"id":"https://openalex.org/T10286","display_name":"Information Retrieval and Search Behavior","score":0.08529999852180481,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.771399974822998},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5350000262260437},{"id":"https://openalex.org/keywords/recall","display_name":"Recall","score":0.5044000148773193},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.4668000042438507},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.46389999985694885},{"id":"https://openalex.org/keywords/coherence","display_name":"Coherence (philosophical gambling strategy)","score":0.4399000108242035},{"id":"https://openalex.org/keywords/context-effect","display_name":"Context effect","score":0.426800012588501}],"concepts":[{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.771399974822998},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6164000034332275},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5350000262260437},{"id":"https://openalex.org/C100660578","wikidata":"https://www.wikidata.org/wiki/Q18733","display_name":"Recall","level":2,"score":0.5044000148773193},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.4668000042438507},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.46389999985694885},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4449000060558319},{"id":"https://openalex.org/C2781181686","wikidata":"https://www.wikidata.org/wiki/Q4226068","display_name":"Coherence (philosophical gambling strategy)","level":2,"score":0.4399000108242035},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4316999912261963},{"id":"https://openalex.org/C76188268","wikidata":"https://www.wikidata.org/wiki/Q1783165","display_name":"Context effect","level":3,"score":0.426800012588501},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.39640000462532043},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.384799987077713},{"id":"https://openalex.org/C64754055","wikidata":"https://www.wikidata.org/wiki/Q7574053","display_name":"Spatial contextual awareness","level":2,"score":0.35850000381469727},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3301999866962433},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3061999976634979},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.3034999966621399},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.3018999993801117},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.2913999855518341},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2912999987602234},{"id":"https://openalex.org/C81669768","wikidata":"https://www.wikidata.org/wiki/Q2359161","display_name":"Precision and recall","level":2,"score":0.28450000286102295},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2687999904155731},{"id":"https://openalex.org/C52085439","wikidata":"https://www.wikidata.org/wiki/Q5165173","display_name":"Context analysis","level":3,"score":0.25839999318122864}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3816713.3819507","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3816713.3819507","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 14th International Conference on Advances in Information Technology","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3816713.3819507","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3816713.3819507","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 14th International Conference on Advances in Information Technology","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":4,"referenced_works":["https://openalex.org/W3099700870","https://openalex.org/W4402670423","https://openalex.org/W4416035738","https://openalex.org/W7131097871"],"related_works":[],"abstract_inverted_index":{"Hallucination":[0],"remains":[1],"a":[2,59,103,117,122,218,264],"major":[3],"challenge":[4],"in":[5,29,279],"Retrieval-Augmented":[6],"Generation":[7],"(RAG),":[8],"especially":[9],"for":[10],"domain-specific":[11],"question":[12],"answering":[13],"over":[14,153,165,212],"heterogeneous":[15],"documents":[16],"containing":[17],"text,":[18],"tables,":[19],"figures,":[20],"and":[21,41,63,81,109,132,140,148,160,198,223,236,263,270,293],"OCR-derived":[22],"content.":[23],"This":[24],"paper":[25],"argues":[26],"that":[27,67,181,229,276],"hallucination":[28,145,277],"document-grounded":[30],"RAG":[31,119,280],"is":[32,114],"strongly":[33],"influenced":[34],"by":[35,246],"context":[36,64,83,100,158,161,199,238],"construction:":[37],"the":[38,44,49,94,97,166,171,182,213,248],"quality,":[39],"structure,":[40],"coherence":[42],"of":[43],"retrieved":[45,89],"evidence":[46,224,250,295],"supplied":[47],"to":[48,93,143,190,196,203,210],"generator.":[50],"To":[51],"address":[52],"this":[53,55],"problem,":[54],"work":[56],"proposes":[57],"L-DSRAG,":[58],"constraint-driven":[60,82],"cross-modal":[61,79,241],"retrieval":[62,221,235],"construction":[65,101,296],"framework":[66,98,253],"integrates":[68],"structure-aware":[69,234],"chunking,":[70],"modality":[71],"normalization,":[72],"hybrid":[73],"retrieval,":[74],"graph-based":[75],"expansion,":[76],"table-aware":[77,237],"grouping,":[78],"pruning,":[80],"packing.":[84],"Rather":[85],"than":[86],"passing":[87],"independently":[88],"top-k":[90],"chunks":[91],"directly":[92],"language":[95],"model,":[96],"treats":[99],"as":[102],"constrained":[104],"evidence-selection":[105],"problem":[106],"under":[107],"token":[108],"structural":[110],"constraints.":[111],"The":[112,252],"system":[113],"evaluated":[115],"against":[116],"traditional":[118],"baseline":[120],"using":[121],"100-question":[123],"benchmark":[124],"covering":[125],"factoid,":[126],"aggregation,":[127],"table-based,":[128],"temporal,":[129],"multi-span,":[130],"cross-modal,":[131],"negative/unanswerable":[133],"queries.":[134],"Four":[135],"questions":[136,169,173],"are":[137,141,151,163],"intentionally":[138],"unanswerable":[139,172],"used":[142],"evaluate":[144],"avoidance.":[146],"Faithfulness":[147],"answer":[149,192],"relevancy":[150,193],"computed":[152,164],"all":[154],"100":[155],"questions,":[156,216],"while":[157,240],"recall":[159,200],"precision":[162,206],"96":[167,214],"answerable":[168,215],"because":[170],"contain":[174],"no":[175],"retrievable":[176],"ground-truth":[177],"evidence.":[178],"Results":[179],"show":[180,228,275],"full":[183],"proposed":[184],"method":[185],"improves":[186],"faithfulness":[187],"from":[188,194,201,208,233],"0.750":[189],"0.900,":[191],"0.740":[195,202],"0.880,":[197],"0.947.":[204],"Context":[205],"decreases":[207],"0.737":[209],"0.693":[211],"reflecting":[217],"tradeoff":[219],"between":[220],"selectivity":[222],"completeness.":[225],"Ablation":[226],"results":[227],"most":[230],"gains":[231],"come":[232],"construction,":[239],"pruning":[242],"provides":[243],"additional":[244],"improvements":[245],"refining":[247],"final":[249],"context.":[251],"operates":[254],"entirely":[255],"with":[256],"locally":[257],"hosted":[258],"components,":[259],"including":[260],"BAAI/bge-large-en-v1.5":[261],"embeddings":[262],"local":[265],"Qwen3-14B":[266],"generator,":[267],"supporting":[268],"privacy-preserving":[269],"API-independent":[271],"deployment.":[272],"These":[273],"findings":[274],"reduction":[278],"can":[281],"be":[282],"achieved":[283],"not":[284],"only":[285],"through":[286,290],"stronger":[287],"generators,":[288],"but":[289],"more":[291],"structured":[292],"selective":[294],"before":[297],"generation.":[298]},"counts_by_year":[],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2026-06-16T00:00:00"}