{"id":"https://openalex.org/W4415342222","doi":"https://doi.org/10.1109/iccv51701.2025.02009","title":"Hallucinatory Image Tokens: A Training-Free EAZY Approach to Detecting and Mitigating Object Hallucinations in LVLMs","display_name":"Hallucinatory Image Tokens: A Training-Free EAZY Approach to Detecting and Mitigating Object Hallucinations in LVLMs","publication_year":2025,"publication_date":"2025-10-01","ids":{"openalex":"https://openalex.org/W4415342222","doi":"https://doi.org/10.1109/iccv51701.2025.02009"},"language":"en","primary_location":{"id":"doi:10.1109/iccv51701.2025.02009","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.02009","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2503.07772","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5086818309","display_name":"Liwei Che","orcid":"https://orcid.org/0000-0002-3741-1087"},"institutions":[{"id":"https://openalex.org/I4210096112","display_name":"Rutgers Sexual and Reproductive Health and Rights","ror":"https://ror.org/00rcvgx40","country_code":"NL","type":"other","lineage":["https://openalex.org/I4210096112"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Liwei Che","raw_affiliation_strings":["Rutgers University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rutgers University","institution_ids":["https://openalex.org/I4210096112"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Tony Qingze Liu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210096112","display_name":"Rutgers Sexual and Reproductive Health and Rights","ror":"https://ror.org/00rcvgx40","country_code":"NL","type":"other","lineage":["https://openalex.org/I4210096112"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Tony Qingze Liu","raw_affiliation_strings":["Rutgers University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rutgers University","institution_ids":["https://openalex.org/I4210096112"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100781888","display_name":"Jing Jia","orcid":"https://orcid.org/0000-0002-3039-0949"},"institutions":[{"id":"https://openalex.org/I4210096112","display_name":"Rutgers Sexual and Reproductive Health and Rights","ror":"https://ror.org/00rcvgx40","country_code":"NL","type":"other","lineage":["https://openalex.org/I4210096112"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Jing Jia","raw_affiliation_strings":["Rutgers University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rutgers University","institution_ids":["https://openalex.org/I4210096112"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5117544622","display_name":"Weiyi Qin","orcid":"https://orcid.org/0009-0001-9225-9188"},"institutions":[{"id":"https://openalex.org/I4210096112","display_name":"Rutgers Sexual and Reproductive Health and Rights","ror":"https://ror.org/00rcvgx40","country_code":"NL","type":"other","lineage":["https://openalex.org/I4210096112"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Weiyi Qin","raw_affiliation_strings":["Rutgers University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rutgers University","institution_ids":["https://openalex.org/I4210096112"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101660252","display_name":"Ruixiang Tang","orcid":"https://orcid.org/0000-0002-5112-8771"},"institutions":[{"id":"https://openalex.org/I4210096112","display_name":"Rutgers Sexual and Reproductive Health and Rights","ror":"https://ror.org/00rcvgx40","country_code":"NL","type":"other","lineage":["https://openalex.org/I4210096112"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Ruixiang Tang","raw_affiliation_strings":["Rutgers University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rutgers University","institution_ids":["https://openalex.org/I4210096112"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5020368096","display_name":"Vladimir Pavlovi\u0107","orcid":"https://orcid.org/0000-0003-3979-1236"},"institutions":[{"id":"https://openalex.org/I4210096112","display_name":"Rutgers Sexual and Reproductive Health and Rights","ror":"https://ror.org/00rcvgx40","country_code":"NL","type":"other","lineage":["https://openalex.org/I4210096112"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Vladimir Pavlovic","raw_affiliation_strings":["Rutgers University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Rutgers University","institution_ids":["https://openalex.org/I4210096112"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210096112"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.31380112,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"21635","last_page":"21644"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12357","display_name":"Digital Media Forensic Detection","score":0.9717000126838684,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.9717000126838684,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.9605000019073486,"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/T12702","display_name":"Brain Tumor Detection and Classification","score":0.954800009727478,"subfield":{"id":"https://openalex.org/subfields/2808","display_name":"Neurology"},"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/focus","display_name":"Focus (optics)","score":0.7688000202178955},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.7512999773025513},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.6974999904632568},{"id":"https://openalex.org/keywords/face","display_name":"Face (sociological concept)","score":0.46129998564720154},{"id":"https://openalex.org/keywords/image-editing","display_name":"Image editing","score":0.41670000553131104},{"id":"https://openalex.org/keywords/hallucinating","display_name":"Hallucinating","score":0.35510000586509705},{"id":"https://openalex.org/keywords/background-image","display_name":"Background image","score":0.33329999446868896}],"concepts":[{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.7688000202178955},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.7512999773025513},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7231000065803528},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7088000178337097},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.6974999904632568},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.5627999901771545},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.46129998564720154},{"id":"https://openalex.org/C2776674983","wikidata":"https://www.wikidata.org/wiki/Q545981","display_name":"Image editing","level":3,"score":0.41670000553131104},{"id":"https://openalex.org/C2911011789","wikidata":"https://www.wikidata.org/wiki/Q130741","display_name":"Hallucinating","level":2,"score":0.35510000586509705},{"id":"https://openalex.org/C3019635856","wikidata":"https://www.wikidata.org/wiki/Q1619726","display_name":"Background image","level":3,"score":0.33329999446868896},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.33219999074935913},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3133000135421753},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.3000999987125397},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.2784999907016754},{"id":"https://openalex.org/C106430172","wikidata":"https://www.wikidata.org/wiki/Q6002272","display_name":"Image restoration","level":4,"score":0.27630001306533813},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2689000070095062},{"id":"https://openalex.org/C64876066","wikidata":"https://www.wikidata.org/wiki/Q5141226","display_name":"Cognitive neuroscience of visual object recognition","level":3,"score":0.26409998536109924}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/iccv51701.2025.02009","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.02009","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2503.07772","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.07772","pdf_url":"https://arxiv.org/pdf/2503.07772","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,"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.2503.07772","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2503.07772","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,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2503.07772","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.07772","pdf_url":"https://arxiv.org/pdf/2503.07772","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,"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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Despite":[0],"their":[1,17],"remarkable":[2,149],"potential,":[3],"Large":[4],"Vision-Language":[5],"Models":[6],"(LVLMs)":[7],"still":[8],"face":[9],"challenges":[10],"with":[11,70],"object":[12,79,136],"hallucination,":[13],"a":[14,61,64,115],"problem":[15],"where":[16],"generated":[18],"outputs":[19],"mistakenly":[20],"incorporate":[21],"objects":[22],"that":[23,119],"do":[24],"not":[25],"actually":[26],"exist.":[27],"Although":[28],"most":[29],"works":[30],"focus":[31,44],"on":[32,109],"addressing":[33],"this":[34,110],"issue":[35,94],"within":[36],"the":[37,43,46,75,93],"language-model":[38],"backbone,":[39],"our":[40],"work":[41],"shifts":[42],"to":[45,56,143,161],"image":[47,53,68,85,91,129],"input":[48],"source,":[49],"investigating":[50],"how":[51],"specific":[52],"tokens":[54,69,86],"contribute":[55],"hallucinations.":[57],"Our":[58],"analysis":[59],"reveals":[60],"striking":[62],"finding:":[63],"small":[65],"subset":[66],"of":[67,78,89],"high":[71],"attention":[72],"scores":[73],"are":[74],"primary":[76],"drivers":[77],"hallucination.":[80],"By":[81],"removing":[82],"these":[83],"hallucinatory":[84],"(only":[87],"1.5%":[88],"all":[90],"tokens),":[92],"can":[95],"be":[96],"effectively":[97],"mitigated.":[98],"This":[99],"finding":[100],"holds":[101],"consistently":[102],"across":[103],"different":[104],"models":[105],"and":[106,122,158],"datasets.":[107],"Building":[108],"insight,":[111],"we":[112],"introduce":[113],"EAZY,":[114],"novel,":[116],"training-free":[117],"method":[118],"automatically":[120],"identifies":[121],"Eliminates":[123],"hAllucinations":[124],"by":[125],"Zeroing":[126],"out":[127],"hallucinatorY":[128],"tokens.":[130],"We":[131],"utilize":[132],"EAZY":[133,147],"for":[134],"unsupervised":[135],"hallucination":[137],"detection,":[138],"achieving":[139],"15%":[140],"improvement":[141],"compared":[142],"previous":[144],"methods.":[145],"Additionally,":[146],"demonstrates":[148],"effectiveness":[150],"in":[151],"mitigating":[152],"hallucinations":[153],"while":[154],"preserving":[155],"model":[156],"utility":[157],"seamlessly":[159],"adapting":[160],"various":[162],"LVLM":[163],"architectures.":[164]},"counts_by_year":[],"updated_date":"2026-07-31T08:31:51.225901","created_date":"2025-10-20T00:00:00"}