{"id":"https://openalex.org/W7164806332","doi":"https://doi.org/10.1145/3805622.3810847","title":"Mitigating Hallucinations in Vision-Language Models via Contextual Entropy Calibration","display_name":"Mitigating Hallucinations in Vision-Language Models via Contextual Entropy Calibration","publication_year":2026,"publication_date":"2026-06-15","ids":{"openalex":"https://openalex.org/W7164806332","doi":"https://doi.org/10.1145/3805622.3810847"},"language":null,"primary_location":{"id":"doi:10.1145/3805622.3810847","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805622.3810847","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 2026 International Conference on Multimedia Retrieval","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3805622.3810847","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138661892","display_name":"Xuanyu Yin","orcid":"https://orcid.org/0009-0002-0049-6328"},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuanyu Yin","raw_affiliation_strings":["Huazhong University of Science and Technology, Wuhan, China"],"raw_orcid":"https://orcid.org/0009-0002-0049-6328","affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology, Wuhan, China","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066682332","display_name":"Daowan Peng","orcid":"https://orcid.org/0000-0001-5208-3811"},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Daowan Peng","raw_affiliation_strings":["Huazhong University of Science and Technology, Wuhan, China"],"raw_orcid":"https://orcid.org/0000-0001-5208-3811","affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology, Wuhan, China","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100323842","display_name":"Wei Wei","orcid":"https://orcid.org/0000-0003-4488-0102"},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei Wei","raw_affiliation_strings":["Huazhong University of Science and Technology, Wuhan, China"],"raw_orcid":"https://orcid.org/0000-0003-4488-0102","affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology, Wuhan, China","institution_ids":["https://openalex.org/I47720641"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I47720641"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.78918495,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2022","last_page":"2031"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12553","display_name":"Psychedelics and Drug Studies","score":0.17949999868869781,"subfield":{"id":"https://openalex.org/subfields/3203","display_name":"Clinical Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12553","display_name":"Psychedelics and Drug Studies","score":0.17949999868869781,"subfield":{"id":"https://openalex.org/subfields/3203","display_name":"Clinical Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.094200000166893,"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/T13397","display_name":"Hallucinations in medical conditions","score":0.04610000178217888,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.4959000051021576},{"id":"https://openalex.org/keywords/entropy","display_name":"Entropy (arrow of time)","score":0.47929999232292175},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.31869998574256897},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.2806999981403351},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.27000001072883606}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5214999914169312},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.4959000051021576},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.49059998989105225},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.47929999232292175},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3370000123977661},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3278000056743622},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.31869998574256897},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.2806999981403351},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.27070000767707825},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.27000001072883606},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2689000070095062},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2565000057220459}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3805622.3810847","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805622.3810847","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 2026 International Conference on Multimedia Retrieval","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3805622.3810847","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805622.3810847","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 2026 International Conference on Multimedia Retrieval","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","score":0.6022539138793945,"display_name":"Reduced inequalities"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W2962735233","https://openalex.org/W4309674289","https://openalex.org/W4322730934","https://openalex.org/W4323338370","https://openalex.org/W4389523832","https://openalex.org/W4389524547","https://openalex.org/W4400647053","https://openalex.org/W4400810332","https://openalex.org/W4401043235","https://openalex.org/W4401414574","https://openalex.org/W4402582321","https://openalex.org/W4402670231","https://openalex.org/W4402670859","https://openalex.org/W4402727764","https://openalex.org/W4402727851","https://openalex.org/W4402753774","https://openalex.org/W4404569270","https://openalex.org/W4404783306","https://openalex.org/W4405596328","https://openalex.org/W4411631896","https://openalex.org/W4411631991","https://openalex.org/W4411635711","https://openalex.org/W4413144618","https://openalex.org/W4413146946","https://openalex.org/W7133193597","https://openalex.org/W7133220561"],"related_works":[],"abstract_inverted_index":{"Large":[0],"Vision-Language":[1],"Models":[2],"(LVLMs)":[3],"have":[4],"shown":[5],"impressive":[6],"performances":[7],"on":[8,102,173,178,188],"multimodal":[9],"tasks,":[10],"yet":[11],"remain":[12],"fundamentally":[13],"constrained":[14],"by":[15],"object":[16,200],"hallucination,":[17],"i.e.,":[18],"generating":[19],"response":[20],"inconsistent":[21],"with":[22,166],"visual":[23,104],"input.":[24],"To":[25],"investigate":[26],"its":[27,203],"underlying":[28],"causes,":[29],"we":[30,57,113],"conduct":[31],"an":[32],"empirical":[33],"analysis":[34],"of":[35,73,110,171,195,209],"internal":[36],"attention":[37,48,60,74,138],"distribution":[38,49],"in":[39,47,198],"various":[40],"LVLMs":[41,210],"and":[42,83,92,155,175],"reveal":[43],"that":[44,123],"the":[45,59,71,193,207],"bias":[46],"imbalance":[50],"varies":[51],"across":[52,76,139],"different":[53,77],"architectural":[54],"models.":[55],"Specifically,":[56],"introduce":[58],"look-back":[61],"rate":[62],"to":[63,99,131,183,205],"design":[64],"a":[65,96,119,167],"contextual":[66,126],"entropy":[67,127],"metric,":[68],"which":[69],"quantifies":[70],"balance":[72],"distributions":[75],"input":[78,140],"components":[79],"(including":[80],"vision,":[81],"instruction":[82],"prior-generated":[84],"response).":[85],"This":[86],"metric":[87],"helps":[88],"distinguish":[89],"between":[90],"hallucinated":[91],"non-hallucinated":[93],"tokens,":[94],"offering":[95],"model-agnostic":[97],"method":[98,197],"detect":[100],"over-reliance":[101],"either":[103],"or":[105,163],"text":[106],"modalities.":[107],"In":[108],"light":[109],"this":[111],"insight,":[112],"propose":[114],"Contextual":[115],"Entropy":[116],"Calibration":[117],"(CEC),":[118],"training-free":[120],"decoding":[121],"strategy":[122],"dynamically":[124],"incorporates":[125],"during":[128],"beam":[129],"search":[130],"steer":[132],"token":[133],"selection":[134],"toward":[135],"more":[136],"balanced":[137],"components.":[141],"Unlike":[142],"existing":[143],"approaches,":[144],"CEC":[145],"divides":[146],"inputs":[147],"at":[148],"each":[149],"generation":[150],"step":[151],"into":[152],"three":[153],"parts":[154],"synthesizes":[156],"their":[157],"effects,":[158],"requiring":[159],"no":[160],"extra":[161],"training":[162],"external":[164],"models,":[165],"maximum":[168],"socre":[169],"improvement":[170],"1.5%":[172],"AMBER":[174],"79":[176],"points":[177],"MME":[179],"perception":[180],"Benchmark":[181],"compared":[182],"previous":[184],"methods.":[185],"Extensive":[186],"experiments":[187],"multiple":[189],"hallucination":[190],"benchmarks":[191],"demonstrate":[192],"effectiveness":[194],"our":[196],"mitigating":[199],"hallucinations,":[201],"highlighting":[202],"potential":[204],"improve":[206],"reliability":[208],"generation.":[211]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-06-16T00:00:00"}