{"id":"https://openalex.org/W4412376969","doi":"https://doi.org/10.1145/3726302.3730081","title":"Seeing Beyond Hallucinations: LLM-based Compositional Information Extraction for Multimodal Reasoning","display_name":"Seeing Beyond Hallucinations: LLM-based Compositional Information Extraction for Multimodal Reasoning","publication_year":2025,"publication_date":"2025-07-13","ids":{"openalex":"https://openalex.org/W4412376969","doi":"https://doi.org/10.1145/3726302.3730081"},"language":"en","primary_location":{"id":"doi:10.1145/3726302.3730081","doi":"https://doi.org/10.1145/3726302.3730081","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3726302.3730081","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3726302.3730081","source":{"id":"https://openalex.org/S4363608773","display_name":"International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3726302.3730081","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103325759","display_name":"Xinwei Li","orcid":"https://orcid.org/0009-0008-0147-858X","observed_orcids":["https://orcid.org/0009-0008-0147-858X"]},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinwei Li","raw_affiliation_strings":["Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0008-0147-858X","affiliations":[{"raw_affiliation_string":"Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Li Lin","orcid":"https://orcid.org/0000-0002-3511-5559","observed_orcids":["https://orcid.org/0000-0002-3511-5559"]},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li Lin","raw_affiliation_strings":["Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-3511-5559","affiliations":[{"raw_affiliation_string":"Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100328284","display_name":"Shuai Wang","orcid":"https://orcid.org/0000-0002-3609-2205","observed_orcids":["https://orcid.org/0000-0002-3609-2205","https://orcid.org/0000-0001-6838-1151"]},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shuai Wang","raw_affiliation_strings":["Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-3609-2205","affiliations":[{"raw_affiliation_string":"Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5055224811","display_name":"Hanqian Wu","orcid":"https://orcid.org/0000-0002-1215-8919","observed_orcids":["https://orcid.org/0000-0002-1215-8919"]},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hanqian Wu","raw_affiliation_strings":["Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-1215-8919","affiliations":[{"raw_affiliation_string":"Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I76569877"],"apc_list":null,"apc_paid":null,"fwci":0.7848,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.71654752,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":93,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"1000","last_page":"1010"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.0,"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/visual-hallucinations","display_name":"visual hallucinations","score":0.6650000214576721},{"id":"https://openalex.org/keywords/multimodal-large-language-models","display_name":"multimodal large language models","score":0.9660000205039978},{"id":"https://openalex.org/keywords/contrastive-decoding","display_name":"contrastive decoding","score":0.8399999737739563},{"id":"https://openalex.org/keywords/hallucination-mitigation","display_name":"hallucination mitigation","score":0.7129999995231628}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6729217767715454},{"id":"https://openalex.org/C195807954","wikidata":"https://www.wikidata.org/wiki/Q1662562","display_name":"Information extraction","level":2,"score":0.4860828220844269},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4763365387916565},{"id":"https://openalex.org/C4725764","wikidata":"https://www.wikidata.org/wiki/Q844704","display_name":"Extraction (chemistry)","level":2,"score":0.4475312829017639},{"id":"https://openalex.org/C4441509","wikidata":"https://www.wikidata.org/wiki/Q6418787","display_name":"Multimodal therapy","level":2,"score":0.4120294153690338},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3699163794517517},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.1809076964855194},{"id":"https://openalex.org/C542102704","wikidata":"https://www.wikidata.org/wiki/Q183257","display_name":"Psychotherapist","level":1,"score":0.1027190089225769},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.06177020072937012},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3726302.3730081","doi":"https://doi.org/10.1145/3726302.3730081","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3726302.3730081","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3726302.3730081","source":{"id":"https://openalex.org/S4363608773","display_name":"International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3726302.3730081","doi":"https://doi.org/10.1145/3726302.3730081","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3726302.3730081","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3726302.3730081","source":{"id":"https://openalex.org/S4363608773","display_name":"International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval","raw_type":"proceedings-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G1198517819","display_name":null,"funder_award_id":"62272098","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3183696597","display_name":null,"funder_award_id":"U24B20152","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G4476335453","display_name":null,"funder_award_id":"62402105","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G7150692672","display_name":null,"funder_award_id":"No.62402105, No. 62272098, No.U24B20152","funder_id":"https://openalex.org/F4320323817","funder_display_name":"Universitas Brawijaya"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320323817","display_name":"Universitas Brawijaya","ror":"https://ror.org/01wk3d929"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412376969.pdf","grobid_xml":"https://content.openalex.org/works/W4412376969.grobid-xml"},"referenced_works_count":43,"referenced_works":["https://openalex.org/W2069065514","https://openalex.org/W2277195237","https://openalex.org/W2463267937","https://openalex.org/W2560730294","https://openalex.org/W2606555609","https://openalex.org/W2892087592","https://openalex.org/W2950104027","https://openalex.org/W2990138404","https://openalex.org/W2996388127","https://openalex.org/W3201174429","https://openalex.org/W3202965415","https://openalex.org/W4306820534","https://openalex.org/W4308713560","https://openalex.org/W4312261477","https://openalex.org/W4312910992","https://openalex.org/W4366330503","https://openalex.org/W4366851068","https://openalex.org/W4383860112","https://openalex.org/W4384642746","https://openalex.org/W4384643636","https://openalex.org/W4385571645","https://openalex.org/W4385801138","https://openalex.org/W4386071604","https://openalex.org/W4386076015","https://openalex.org/W4389519483","https://openalex.org/W4389523832","https://openalex.org/W4390874566","https://openalex.org/W4393160204","https://openalex.org/W4395483878","https://openalex.org/W4400530309","https://openalex.org/W4400530533","https://openalex.org/W4400530983","https://openalex.org/W4402716477","https://openalex.org/W4402727624","https://openalex.org/W4402753774","https://openalex.org/W4403081466","https://openalex.org/W4403990523","https://openalex.org/W4404612908","https://openalex.org/W4404782034","https://openalex.org/W4412230188","https://openalex.org/W6600756316","https://openalex.org/W7133189340","https://openalex.org/W7133193597"],"related_works":["https://openalex.org/W4292002379","https://openalex.org/W2358901620","https://openalex.org/W4306253668","https://openalex.org/W3032782782","https://openalex.org/W2377297411","https://openalex.org/W3148217948","https://openalex.org/W2413547588","https://openalex.org/W1991896163","https://openalex.org/W3204019825","https://openalex.org/W2368651715"],"abstract_inverted_index":{"Advancements":[0],"in":[1,38,143],"Multimodal":[2,91],"Large":[3],"Language":[4],"Models":[5],"(MLLMs)":[6],"have":[7],"significantly":[8],"improved":[9],"information":[10],"extraction":[11,57],"and":[12,51,65,81,115,123,151,160,179,181,212],"retrieval":[13],"performance.":[14],"Despite":[15],"these":[16,71,87,155],"achievements,":[17],"MLLMs":[18,120],"still":[19],"suffer":[20],"from":[21,45],"the":[22,39,76,82,117,131,157,208],"visual":[23,59],"object":[24,109,183],"hallucination":[25],"problem,":[26],"where":[27],"models":[28],"produce":[29],"plausible,":[30],"yet":[31],"incorrect,":[32],"or":[33,130],"irrelevant":[34],"content":[35],"not":[36],"present":[37],"input":[40],"data.":[41],"This":[42,136],"issue":[43],"arises":[44],"an":[46],"over-reliance":[47],"on":[48,204],"''bag-of-objects''":[49],"representations":[50],"language":[52,113],"priors,":[53],"leading":[54],"to":[55,69,121,196],"inadequate":[56],"of":[58,119,133,149,176],"objects,":[60],"along":[61],"with":[62,104,156],"their":[63],"attributes":[64,150],"relationships.":[66],"Existing":[67],"methods":[68,193],"mitigate":[70],"hallucinations":[72,110],"are":[73],"limited":[74],"by":[75,112,139],"significant":[77],"human":[78],"labor":[79],"required":[80],"coarse-grained":[83],"nature.":[84],"To":[85],"overcome":[86],"challenges,":[88],"we":[89],"introduce":[90],"Contrastive":[92],"Decoding":[93],"(MMCD),":[94],"a":[95],"novel":[96],"decoding":[97,192],"approach":[98],"that":[99,187],"integrates":[100],"graph-structured":[101],"reasoning":[102,170],"paths":[103],"contrastive":[105],"decoding.":[106],"MMCD":[107,188,200],"mitigates":[108],"induced":[111],"priors":[114],"enhances":[116],"ability":[118],"extract":[122],"understand":[124],"compositional":[125,169],"information,":[126],"without":[127],"additional":[128],"training":[129],"usage":[132],"external":[134],"tools.":[135],"is":[137],"achieved":[138],"masking":[140],"key":[141],"objects":[142],"images,":[144],"constructing":[145],"perturbed":[146],"scene":[147,161],"graphs":[148],"relationships,":[152],"then":[153],"contrasting":[154],"original":[158],"image":[159,178],"graph.":[162],"Extensive":[163],"evaluation":[164],"across":[165],"three":[166],"distinct":[167],"multimodal":[168],"tasks:":[171],"spatial":[172],"relationship":[173],"reasoning,":[174],"alignment":[175],"synthetic":[177],"caption,":[180],"fine-grained":[182],"attribute":[184],"understanding,":[185],"show":[186],"consistently":[189],"surpasses":[190],"existing":[191],"when":[194],"applied":[195],"various":[197],"MLLMs.":[198],"Moreover,":[199],"achieves":[201],"state-of-the-art":[202],"performance":[203],"multiple":[205],"benchmarks,":[206],"including":[207],"What's":[209],"Up,":[210],"SeeTrue":[211],"SugarCrepe":[213],"datasets.":[214]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2025-10-10T00:00:00"}