{"id":"https://openalex.org/W7224254937","doi":"https://doi.org/10.18653/2026.emnlp-main.2272","title":"Mitigating Object Hallucination in MLLMs with Global-to-Local Visual Attention Correction","display_name":"Mitigating Object Hallucination in MLLMs with Global-to-Local Visual Attention Correction","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7224254937","doi":"https://doi.org/10.18653/2026.emnlp-main.2272"},"language":"en","primary_location":{"id":"doi:10.18653/2026.emnlp-main.2272","doi":"https://doi.org/10.18653/2026.emnlp-main.2272","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.2272","pdf_url":"https://aclanthology.org/2026.emnlp-main.2272.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","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 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://aclanthology.org/2026.emnlp-main.2272.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5159908572","display_name":"Xingxing Huang","orcid":"https://orcid.org/0000-0002-7043-9794","observed_orcids":["https://orcid.org/0000-0002-7043-9794"]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xingxing Huang","raw_affiliation_strings":["School of Computer Science , USTC , China;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science , USTC , China;","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160073503","display_name":"Zhiguo Yang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhiguo Yang","raw_affiliation_strings":["School of Computer Science , USTC , China;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science , USTC , China;","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008872086","display_name":"Jiacheng Pi","orcid":"https://orcid.org/0009-0007-8024-9891","observed_orcids":["https://orcid.org/0009-0007-8024-9891"]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiacheng Pi","raw_affiliation_strings":["School of Computer Science , USTC , China;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science , USTC , China;","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160120029","display_name":"Dongsheng Xu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dongsheng Xu","raw_affiliation_strings":["School of Computer Science , USTC , China;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science , USTC , China;","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160069674","display_name":"Ruizhi Zhong","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ruizhi Zhong","raw_affiliation_strings":["School of Computer Science , USTC , China;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science , USTC , China;","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028614556","display_name":"Peipei Xu","orcid":"https://orcid.org/0000-0002-1781-3820","observed_orcids":["https://orcid.org/0000-0002-1781-3820"]},"institutions":[{"id":"https://openalex.org/I177639307","display_name":"University of Dundee","ror":"https://ror.org/03h2bxq36","country_code":"GB","type":"education","lineage":["https://openalex.org/I177639307"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Peipei Xu","raw_affiliation_strings":["University of Dundee , UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Dundee , UK","institution_ids":["https://openalex.org/I177639307"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5074225885","display_name":"Wenjie Ruan","orcid":"https://orcid.org/0000-0002-8311-8738","observed_orcids":["https://orcid.org/0000-0002-8311-8738"]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenjie Ruan","raw_affiliation_strings":["School of Computer Science , USTC , China;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science , USTC , China;","institution_ids":["https://openalex.org/I126520041"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.9111913,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"51671","last_page":"51689"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9990000128746033,"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":0.9990000128746033,"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":0.00019999999494757503,"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/T10028","display_name":"Topic Modeling","score":0.00019999999494757503,"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/object-hallucination","display_name":"object hallucination","score":0.9750000238418579},{"id":"https://openalex.org/keywords/multimodal-large-language-models","display_name":"multimodal large language models","score":0.9490000009536743},{"id":"https://openalex.org/keywords/visual-attention","display_name":"visual attention","score":0.8809999823570251},{"id":"https://openalex.org/keywords/hallucination-mitigation","display_name":"hallucination mitigation","score":0.7789999842643738},{"id":"https://openalex.org/keywords/vision-language-models","display_name":"vision-language models","score":0.6880000233650208}],"concepts":[{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.6197999715805054},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.5648999810218811},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.557200014591217},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5162000060081482},{"id":"https://openalex.org/C2986089797","wikidata":"https://www.wikidata.org/wiki/Q6501338","display_name":"Visual attention","level":3,"score":0.5138000249862671},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.40290001034736633},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.3702000081539154},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3231000006198883},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.310699999332428},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.2856000065803528},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.27559998631477356},{"id":"https://openalex.org/C56461940","wikidata":"https://www.wikidata.org/wiki/Q970687","display_name":"Eye tracking","level":2,"score":0.26499998569488525},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.26440000534057617},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C46312422","wikidata":"https://www.wikidata.org/wiki/Q11024","display_name":"Communication","level":1,"score":0.25429999828338623},{"id":"https://openalex.org/C158495155","wikidata":"https://www.wikidata.org/wiki/Q2369151","display_name":"Visual search","level":2,"score":0.2515000104904175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/2026.emnlp-main.2272","doi":"https://doi.org/10.18653/2026.emnlp-main.2272","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.2272","pdf_url":"https://aclanthology.org/2026.emnlp-main.2272.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","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 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/2026.emnlp-main.2272","doi":"https://doi.org/10.18653/2026.emnlp-main.2272","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.2272","pdf_url":"https://aclanthology.org/2026.emnlp-main.2272.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","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 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7224254937.pdf","grobid_xml":"https://content.openalex.org/works/W7224254937.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7224326178","https://openalex.org/W7224518630","https://openalex.org/W7155615112","https://openalex.org/W7221035416","https://openalex.org/W7224278437","https://openalex.org/W7166892059","https://openalex.org/W7220898065","https://openalex.org/W7224180379","https://openalex.org/W7140113582","https://openalex.org/W4411119126"],"abstract_inverted_index":{"MultimodalLarge":[0],"Language":[1],"Models":[2],"(MLLMs)":[3],"have":[4],"achieved":[5],"impressive":[6],"progress":[7],"in":[8,29,40,57,88],"vision-language":[9],"understanding,":[10],"yet":[11],"they":[12],"remain":[13],"susceptible":[14],"to":[15,20,123],"object":[16,147],"hallucination,":[17],"which":[18],"refers":[19],"generating":[21],"descriptions":[22],"of":[23,61],"objects":[24],"that":[25,55,80],"do":[26],"not":[27],"exist":[28],"the":[30,58,62,103,125],"input":[31],"image.Through":[32],"comprehensive":[33],"analysis,":[34],"we":[35,68],"uncover":[36],"a":[37,75],"fundamental":[38],"trade-off":[39],"MLLMs:":[41],"strengthening":[42],"visual":[43,89,100],"attention":[44],"enhances":[45],"grounding":[46],"completeness":[47,152],"but":[48],"also":[49],"increases":[50],"hallucinations,":[51],"especially":[52],"additive":[53],"hallucinations":[54,148],"appear":[56],"trailing":[59],"part":[60],"generated":[63],"sentence.To":[64],"alleviate":[65],"this":[66],"trade-off,":[67],"propose":[69],"GLVAC,":[70],"Global-to-Local":[71],"Visual":[72],"Attention":[73],"Correction,":[74],"training-free":[76],"dual-path":[77],"inference":[78],"framework":[79],"adaptively":[81],"combines":[82],"global":[83,91],"conservatism":[84],"and":[85,95,130,156],"local":[86,104],"enhancement":[87],"attention.The":[90],"path":[92],"maintains":[93],"stable":[94],"grounded":[96],"predictions":[97],"under":[98],"moderate":[99],"attention,":[101],"while":[102,149],"correction":[105,126],"branch":[106],"is":[107],"dynamically":[108],"activated":[109],"for":[110],"high-uncertainty":[111],"tokens":[112],"identified":[113],"via":[114],"entropy-based":[115],"estimation.A":[116],"confidence-guided":[117],"gating":[118],"mechanism":[119],"then":[120],"determines":[121],"whether":[122],"adopt":[124],"output,":[127],"ensuring":[128],"safe":[129],"reliable":[131],"refinement.GLVAC":[132],"can":[133],"be":[134],"seamlessly":[135],"integrated":[136],"into":[137],"existing":[138],"MLLMs":[139],"without":[140],"retraining":[141],"or":[142],"external":[143],"modules,":[144],"effectively":[145],"reducing":[146],"maintaining":[150],"perceptual":[151],"across":[153],"diverse":[154],"benchmarks":[155],"model":[157],"families.":[158]},"counts_by_year":[],"updated_date":"2026-11-01T11:40:01.658920","created_date":"2026-10-30T00:00:00"}