{"id":"https://openalex.org/W7135011248","doi":"https://doi.org/10.48550/arxiv.2603.10360","title":"One Token, Two Fates: A Unified Framework via Vision Token Manipulation Against MLLMs Hallucination","display_name":"One Token, Two Fates: A Unified Framework via Vision Token Manipulation Against MLLMs Hallucination","publication_year":2026,"publication_date":"2026-03-11","ids":{"openalex":"https://openalex.org/W7135011248","doi":"https://doi.org/10.48550/arxiv.2603.10360"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.10360","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.10360","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":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.10360","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128903671","display_name":"Zhan Fa","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fa, Zhan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128884103","display_name":"Yue Duan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Duan, Yue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128842623","display_name":"Jian Jun Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Jian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128910643","display_name":"Lei Qi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qi, Lei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128864958","display_name":"Yinghuan Shi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shi, Yinghuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.5383999943733215,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.5383999943733215,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.053199999034404755,"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/T13397","display_name":"Hallucinations in medical conditions","score":0.02979999966919422,"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/security-token","display_name":"Security token","score":0.6833999752998352},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5810999870300293},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.5317000150680542},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.5271999835968018},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.48980000615119934},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.3968000113964081},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.3691999912261963},{"id":"https://openalex.org/keywords/machine-vision","display_name":"Machine vision","score":0.34360000491142273}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7840999960899353},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.6833999752998352},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6552000045776367},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5810999870300293},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.5745000243186951},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.5317000150680542},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5271999835968018},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.48980000615119934},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.3968000113964081},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3691999912261963},{"id":"https://openalex.org/C5339829","wikidata":"https://www.wikidata.org/wiki/Q1425977","display_name":"Machine vision","level":2,"score":0.34360000491142273},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.31769999861717224},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3084999918937683},{"id":"https://openalex.org/C64876066","wikidata":"https://www.wikidata.org/wiki/Q5141226","display_name":"Cognitive neuroscience of visual object recognition","level":3,"score":0.30660000443458557},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.30169999599456787},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.2904999852180481},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.29019999504089355},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2696000039577484},{"id":"https://openalex.org/C193611912","wikidata":"https://www.wikidata.org/wiki/Q4677596","display_name":"Active vision","level":2,"score":0.266400009393692},{"id":"https://openalex.org/C160633673","wikidata":"https://www.wikidata.org/wiki/Q355198","display_name":"Pixel","level":2,"score":0.2660999894142151},{"id":"https://openalex.org/C522192633","wikidata":"https://www.wikidata.org/wiki/Q34228","display_name":"Sign language","level":2,"score":0.25600001215934753}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.10360","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.10360","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":"doi:10.48550/arxiv.2603.10360","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.10360","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":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Current":[0],"training-free":[1],"methods":[2,20],"tackle":[3],"MLLM":[4],"hallucination":[5,91],"with":[6,179],"separate":[7,19],"strategies:":[8],"either":[9],"enhancing":[10,28],"visual":[11,82,127],"signals":[12],"or":[13],"suppressing":[14,37],"text":[15],"inertia.":[16],"However,":[17],"these":[18,151],"are":[21],"insufficient":[22],"due":[23],"to":[24,125,139],"critical":[25],"trade-offs:":[26],"simply":[27],"vision":[29,69,87,105],"often":[30],"fails":[31],"against":[32],"strong":[33],"language":[34,38],"prior,":[35],"while":[36,129],"can":[39],"introduce":[40],"extra":[41],"image-irrelevant":[42],"noise.":[43],"Moreover,":[44],"we":[45],"find":[46],"their":[47],"naive":[48],"combination":[49],"is":[50],"also":[51],"ineffective,":[52],"necessitating":[53],"a":[54,60,181],"unified":[55],"framework.":[56],"We":[57],"propose":[58],"such":[59],"framework":[61,103,155],"by":[62,168],"focusing":[63],"on":[64,100,113,174],"the":[65,68,158],"core":[66],"asset:":[67],"token.":[70],"Our":[71],"design":[72],"leverages":[73],"two":[74,108,152],"key":[75],"insights:":[76],"(1)":[77],"augmented":[78,123],"images":[79,97],"offer":[80],"complementary":[81],"semantics,":[83],"and":[84],"(2)":[85],"removing":[86],"tokens":[88,106,124,138],"(information-gap)":[89],"isolates":[90],"tendencies":[92],"more":[93],"precisely":[94],"than":[95],"distorting":[96],"(modality-gap).":[98],"Based":[99],"these,":[101],"our":[102,116,130,154],"uses":[104,136],"in":[107],"distinct":[109],"ways,":[110],"both":[111],"operating":[112],"latent":[114],"representations:":[115],"Synergistic":[117],"Visual":[118],"Calibration":[119,133],"(SVC)":[120],"module":[121,135],"incorporates":[122],"strengthen":[126],"representations,":[128],"Causal":[131],"Representation":[132],"(CRC)":[134],"pruned":[137],"create":[140],"latent-space":[141],"negative":[142],"samples":[143],"for":[144],"correcting":[145],"internal":[146],"model":[147],"biases.":[148],"By":[149],"harmonizing":[150],"roles,":[153],"effectively":[156],"restores":[157],"vision-language":[159],"balance,":[160],"significantly":[161],"reducing":[162],"object":[163],"hallucinations,":[164],"improving":[165],"POPE":[166],"accuracy":[167],"an":[169],"average":[170],"of":[171],"2%":[172],"absolute":[173],"LLaVA-1.5":[175],"across":[176],"multiple":[177],"benchmarks":[178],"only":[180],"1.06x":[182],"inference":[183],"latency":[184],"overhead.":[185]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-03-13T00:00:00"}