{"id":"https://openalex.org/W7171939381","doi":"https://doi.org/10.1145/3774521.3774542","title":"HalluShift++: Bridging Language and Vision through Internal Representation Shifts for Hierarchical Hallucinations in MLLMs","display_name":"HalluShift++: Bridging Language and Vision through Internal Representation Shifts for Hierarchical Hallucinations in MLLMs","publication_year":2025,"publication_date":"2025-12-17","ids":{"openalex":"https://openalex.org/W7171939381","doi":"https://doi.org/10.1145/3774521.3774542"},"language":null,"primary_location":{"id":"doi:10.1145/3774521.3774542","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774521.3774542","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 Sixteen Indian Conference on Computer Vision, Graphics and Image Processing","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3774521.3774542","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5117110863","display_name":"Sujoy Nath","orcid":null},"institutions":[{"id":"https://openalex.org/I36090812","display_name":"Netaji Subhas University of Technology","ror":"https://ror.org/01fczmh85","country_code":"IN","type":"education","lineage":["https://openalex.org/I36090812"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Sujoy Nath","raw_affiliation_strings":["Netaji Subhash Engineering College, Kolkata, India"],"raw_orcid":"https://orcid.org/0009-0003-9974-8018","affiliations":[{"raw_affiliation_string":"Netaji Subhash Engineering College, Kolkata, India","institution_ids":["https://openalex.org/I36090812"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055678857","display_name":"Arkaprabha Basu","orcid":"https://orcid.org/0000-0003-0136-0887"},"institutions":[{"id":"https://openalex.org/I4210159890","display_name":"TCG Crest","ror":"https://ror.org/05h2r8y34","country_code":"IN","type":"other","lineage":["https://openalex.org/I4210159890"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Arkaprabha Basu","raw_affiliation_strings":["TCG Crest, Kolkata, India"],"raw_orcid":"https://orcid.org/0000-0003-0136-0887","affiliations":[{"raw_affiliation_string":"TCG Crest, Kolkata, India","institution_ids":["https://openalex.org/I4210159890"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134415510","display_name":"Sharanya Dasgupta","orcid":null},"institutions":[{"id":"https://openalex.org/I6498739","display_name":"Indian Statistical Institute","ror":"https://ror.org/00q2w1j53","country_code":"IN","type":"education","lineage":["https://openalex.org/I6498739"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Sharanya Dasgupta","raw_affiliation_strings":["Indian Statistical Institute, Kolkata, India"],"raw_orcid":"https://orcid.org/0009-0006-6385-243X","affiliations":[{"raw_affiliation_string":"Indian Statistical Institute, Kolkata, India","institution_ids":["https://openalex.org/I6498739"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5000078546","display_name":"Swagatam Das","orcid":"https://orcid.org/0000-0001-6843-4508"},"institutions":[{"id":"https://openalex.org/I6498739","display_name":"Indian Statistical Institute","ror":"https://ror.org/00q2w1j53","country_code":"IN","type":"education","lineage":["https://openalex.org/I6498739"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Swagatam Das","raw_affiliation_strings":["Electronics and Communication Sciences Unit, Indian Statistical Institute, Kolkata, India"],"raw_orcid":"https://orcid.org/0000-0001-6843-4508","affiliations":[{"raw_affiliation_string":"Electronics and Communication Sciences Unit, Indian Statistical Institute, Kolkata, India","institution_ids":["https://openalex.org/I6498739"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"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":"1","last_page":"9"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":null,"topics":[],"keywords":[{"id":"https://openalex.org/keywords/bridging","display_name":"Bridging (networking)","score":0.7178999781608582},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5551000237464905},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.483599990606308},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.39660000801086426},{"id":"https://openalex.org/keywords/context-model","display_name":"Context model","score":0.36329999566078186},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.3328999876976013}],"concepts":[{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.7178999781608582},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5551000237464905},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.5440999865531921},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5299000144004822},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.483599990606308},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.438400000333786},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.39660000801086426},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.38269999623298645},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.37380000948905945},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.36329999566078186},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3382999897003174},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.3328999876976013},{"id":"https://openalex.org/C2983448237","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Language understanding","level":2,"score":0.33160001039505005},{"id":"https://openalex.org/C51929080","wikidata":"https://www.wikidata.org/wiki/Q2425187","display_name":"Codebase","level":3,"score":0.32850000262260437},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.303600013256073},{"id":"https://openalex.org/C2908998935","wikidata":"https://www.wikidata.org/wiki/Q130741","display_name":"Visual Hallucination","level":2,"score":0.298799991607666},{"id":"https://openalex.org/C2780878386","wikidata":"https://www.wikidata.org/wiki/Q1659648","display_name":"Visual language","level":2,"score":0.2623000144958496}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3774521.3774542","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774521.3774542","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 Sixteen Indian Conference on Computer Vision, Graphics and Image Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3774521.3774542","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3774521.3774542","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 Sixteen Indian Conference on Computer Vision, Graphics and Image Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":8,"referenced_works":["https://openalex.org/W2148143831","https://openalex.org/W2962735233","https://openalex.org/W4309674289","https://openalex.org/W4413146493","https://openalex.org/W4415796039","https://openalex.org/W4415798129","https://openalex.org/W4416250893","https://openalex.org/W7133235880"],"related_works":[],"abstract_inverted_index":{"Multimodal":[0],"Large":[1],"Language":[2],"Models":[3],"(MLLMs)":[4],"have":[5],"demonstrated":[6],"remarkable":[7],"capabilities":[8],"in":[9,46,52,75,108],"vision-language":[10],"understanding":[11],"tasks.":[12],"While":[13],"these":[14],"models":[15,132],"often":[16,22],"produce":[17],"linguistically":[18],"coherent":[19],"output,":[20],"they":[21],"suffer":[23],"from":[24,128],"hallucinations,":[25],"generating":[26],"descriptions":[27],"that":[28,87],"are":[29,66],"factually":[30],"inconsistent":[31],"with":[32],"the":[33,42,53,85,94,109,123],"visual":[34],"content,":[35],"potentially":[36],"leading":[37],"to":[38,69,103,134],"adverse":[39],"consequences.":[40],"Therefore,":[41],"assessment":[43],"of":[44,77,98,111,114,125],"hallucinations":[45,70],"MLLM":[47],"has":[48],"become":[49],"increasingly":[50],"crucial":[51],"model":[54],"development":[55],"process.":[56],"Contemporary":[57],"methodologies":[58],"predominantly":[59],"depend":[60],"on":[61],"external":[62],"LLM":[63],"evaluators,":[64],"which":[65],"themselves":[67],"susceptible":[68],"and":[71],"may":[72],"present":[73],"challenges":[74],"terms":[76],"domain":[78],"adaptation.":[79],"In":[80],"this":[81],"study,":[82],"we":[83],"propose":[84],"hypothesis":[86],"hallucination":[88,126],"manifests":[89],"as":[90],"measurable":[91],"irregularities":[92],"within":[93],"internal":[95],"layer":[96],"dynamics":[97],"MLLMs,":[99],"not":[100],"merely":[101],"due":[102],"distributional":[104],"shifts":[105],"but":[106],"also":[107],"context":[110],"layer-wise":[112],"analysis":[113],"specific":[115],"assumptions.":[116],"By":[117],"incorporating":[118],"such":[119],"modifications,":[120],"HalluShift++":[121],"broadens":[122],"efficacy":[124],"detection":[127],"text-based":[129],"large":[130],"language":[131],"(LLMs)":[133],"encompass":[135],"multimodal":[136],"scenarios.":[137],"Our":[138],"codebase":[139],"is":[140],"available":[141],"at":[142],"https://github.com/C0mRD/HalluShift_Plus.":[143]},"counts_by_year":[],"updated_date":"2026-08-01T09:00:35.917206","created_date":"2026-08-01T00:00:00"}
