{"id":"https://openalex.org/W7135204965","doi":"https://doi.org/10.48550/arxiv.2603.12149","title":"Linking Perception, Confidence and Accuracy in MLLMs","display_name":"Linking Perception, Confidence and Accuracy in MLLMs","publication_year":2026,"publication_date":"2026-03-12","ids":{"openalex":"https://openalex.org/W7135204965","doi":"https://doi.org/10.48550/arxiv.2603.12149"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.12149","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.12149","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.12149","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5108958967","display_name":"Yuetian Du","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Du, Yuetian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128925734","display_name":"Yucheng Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yucheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128943083","display_name":"Rongyu Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Rongyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129009172","display_name":"Zhijie Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Zhijie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128966750","display_name":"Boyu Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Boyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128980665","display_name":"Ming Kong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kong, Ming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128935434","display_name":"Jie Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Jie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129065132","display_name":"Qiang Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Qiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":8,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.7109000086784363,"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":0.7109000086784363,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.1728000044822693,"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":0.02410000003874302,"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/scaling","display_name":"Scaling","score":0.5781000256538391},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.534500002861023},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5270000100135803},{"id":"https://openalex.org/keywords/sensitivity","display_name":"Sensitivity (control systems)","score":0.4708000123500824},{"id":"https://openalex.org/keywords/schedule","display_name":"Schedule","score":0.4542999863624573},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.3837999999523163},{"id":"https://openalex.org/keywords/low-confidence","display_name":"Low Confidence","score":0.3804999887943268}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6746000051498413},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6108999848365784},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.5781000256538391},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.534500002861023},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5270000100135803},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5169000029563904},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.4708000123500824},{"id":"https://openalex.org/C68387754","wikidata":"https://www.wikidata.org/wiki/Q7271585","display_name":"Schedule","level":2,"score":0.4542999863624573},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.3837999999523163},{"id":"https://openalex.org/C2909755999","wikidata":"https://www.wikidata.org/wiki/Q4751126","display_name":"Low Confidence","level":2,"score":0.3804999887943268},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.33709999918937683},{"id":"https://openalex.org/C44249647","wikidata":"https://www.wikidata.org/wiki/Q208498","display_name":"Confidence interval","level":2,"score":0.2773999869823456},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.27399998903274536},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.273499995470047},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.2667999863624573},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.26669999957084656},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.25270000100135803}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.12149","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.12149","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.12149","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.12149","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"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":{"Recent":[0],"advances":[1],"in":[2,43,112],"Multi-modal":[3],"Large":[4],"Language":[5],"Models":[6],"(MLLMs)":[7],"have":[8],"predominantly":[9],"focused":[10],"on":[11],"enhancing":[12],"visual":[13],"perception":[14],"to":[15,64,119],"improve":[16],"accuracy.":[17],"However,":[18],"a":[19,33,38,60,85],"critical":[20],"question":[21],"remains":[22],"unexplored:":[23],"Do":[24],"models":[25],"know":[26],"when":[27],"they":[28],"do":[29],"not":[30],"know?":[31],"Through":[32],"probing":[34],"experiment,":[35],"we":[36,48],"reveal":[37],"severe":[39],"confidence":[40,78,106],"miscalibration":[41],"problem":[42],"MLLMs.":[44],"To":[45],"address":[46],"this,":[47],"propose":[49,90],"Confidence-Driven":[50],"Reinforcement":[51],"Learning":[52],"(CDRL),":[53],"which":[54,95],"uses":[55],"original-noise":[56],"image":[57],"pairs":[58],"and":[59,68,100,123,150],"novel":[61],"confidence-based":[62],"reward":[63],"enhance":[65],"perceptual":[66],"sensitivity":[67],"robustly":[69],"calibrate":[70],"the":[71,145],"model's":[72],"confidence.":[73],"Beyond":[74],"training":[75],"benefits,":[76],"calibrated":[77],"enables":[79],"more":[80],"effective":[81],"test-time":[82],"scaling":[83,151],"as":[84],"free":[86],"lunch.":[87],"We":[88],"further":[89],"Confidence-Aware":[91],"Test-Time":[92],"Scaling":[93],"(CA-TTS),":[94],"dynamically":[96],"coordinates":[97],"Self-Consistency,":[98],"Self-Reflection,":[99],"Visual":[101],"Self-Check":[102],"modules":[103,122],"guided":[104],"by":[105],"signals.":[107],"An":[108],"Expert":[109],"Model":[110],"acts":[111],"multiple":[113],"roles":[114],"(e.g.,":[115],"Planner,":[116],"Critic,":[117],"Voter)":[118],"schedule":[120],"these":[121],"provide":[124],"external":[125],"verification.":[126],"Our":[127],"integrated":[128],"framework":[129],"establishes":[130],"new":[131],"state-of-the-art":[132],"results":[133],"with":[134],"consistent":[135],"8.8%":[136],"gains":[137],"across":[138],"four":[139],"benchmarks.":[140],"More":[141],"ablation":[142],"studies":[143],"demonstrate":[144],"effectiveness":[146],"of":[147],"each":[148],"module":[149],"superiority.":[152]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-14T00:00:00"}