{"id":"https://openalex.org/W4417069263","doi":"https://doi.org/10.1109/iccv51701.2025.00391","title":"Why LVLMs are More Prone to Hallucinations in Longer Responses: The Role of Context","display_name":"Why LVLMs are More Prone to Hallucinations in Longer Responses: The Role of Context","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4417069263","doi":"https://doi.org/10.1109/iccv51701.2025.00391"},"language":null,"primary_location":{"id":"doi:10.1109/iccv51701.2025.00391","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.00391","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":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2510.20229","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5061928568","display_name":"Zheng Ge","orcid":"https://orcid.org/0000-0002-9983-7120"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ge Zheng","raw_affiliation_strings":["School of Computer Science and Engineering, Sun Yat-sen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Sun Yat-sen University","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jiaye Qian","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiaye Qian","raw_affiliation_strings":["ShanghaiTech University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ShanghaiTech University","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102705973","display_name":"Jiajin Tang","orcid":"https://orcid.org/0009-0002-2906-8941"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiajin Tang","raw_affiliation_strings":["ShanghaiTech University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ShanghaiTech University","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5012166205","display_name":"Sibei Yang","orcid":"https://orcid.org/0000-0002-8144-7351"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sibei Yang","raw_affiliation_strings":["School of Computer Science and Engineering, Sun Yat-sen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering, Sun Yat-sen University","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.1987198,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4101","last_page":"4113"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.11840000003576279,"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":0.11840000003576279,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.08799999952316284,"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/T11094","display_name":"Face Recognition and Perception","score":0.05429999902844429,"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/context","display_name":"Context (archaeology)","score":0.6845999956130981},{"id":"https://openalex.org/keywords/coherence","display_name":"Coherence (philosophical gambling strategy)","score":0.45559999346733093},{"id":"https://openalex.org/keywords/auditory-hallucination","display_name":"Auditory hallucination","score":0.4185999929904938},{"id":"https://openalex.org/keywords/visual-hallucination","display_name":"Visual Hallucination","score":0.367000013589859},{"id":"https://openalex.org/keywords/completeness","display_name":"Completeness (order theory)","score":0.3366999924182892},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.28049999475479126}],"concepts":[{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6845999956130981},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.6446999907493591},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.6381000280380249},{"id":"https://openalex.org/C2781181686","wikidata":"https://www.wikidata.org/wiki/Q4226068","display_name":"Coherence (philosophical gambling strategy)","level":2,"score":0.45559999346733093},{"id":"https://openalex.org/C2776706361","wikidata":"https://www.wikidata.org/wiki/Q1054088","display_name":"Auditory hallucination","level":3,"score":0.4185999929904938},{"id":"https://openalex.org/C2908998935","wikidata":"https://www.wikidata.org/wiki/Q130741","display_name":"Visual Hallucination","level":2,"score":0.367000013589859},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3443000018596649},{"id":"https://openalex.org/C17231256","wikidata":"https://www.wikidata.org/wiki/Q5156540","display_name":"Completeness (order theory)","level":2,"score":0.3366999924182892},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.28209999203681946},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.28049999475479126},{"id":"https://openalex.org/C2779043874","wikidata":"https://www.wikidata.org/wiki/Q189643","display_name":"Delusion","level":2,"score":0.26570001244544983},{"id":"https://openalex.org/C76188268","wikidata":"https://www.wikidata.org/wiki/Q1783165","display_name":"Context effect","level":3,"score":0.2630999982357025},{"id":"https://openalex.org/C2779727114","wikidata":"https://www.wikidata.org/wiki/Q170082","display_name":"Psychosis","level":2,"score":0.2578999996185303},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.25369998812675476},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.25110000371932983}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/iccv51701.2025.00391","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.00391","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:2510.20229","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.20229","pdf_url":"https://arxiv.org/pdf/2510.20229","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2510.20229","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.20229","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":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2510.20229","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.20229","pdf_url":"https://arxiv.org/pdf/2510.20229","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2942363109","display_name":null,"funder_award_id":"62206174","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Vision-Language":[1],"Models":[2],"(LVLMs)":[3],"have":[4],"made":[5],"significant":[6,126],"progress":[7],"in":[8,22,83,179],"recent":[9],"years":[10],"but":[11,72],"are":[12],"also":[13],"prone":[14],"to":[15,28,163],"hallucination":[16,38,139],"issues.":[17],"They":[18],"exhibit":[19],"more":[20,147],"hallucinations":[21,65,99,118,178],"longer,":[23],"free-form":[24],"responses,":[25],"often":[26],"attributed":[27],"accumulated":[29],"uncertainties.":[30],"In":[31],"this":[32,160],"paper,":[33],"we":[34,59,90],"ask:":[35],"Does":[36],"increased":[37,75],"result":[39],"solely":[40,156],"from":[41],"length-induced":[42],"errors,":[43],"or":[44],"is":[45,66],"there":[46],"a":[47,52,92,170,174],"deeper":[48,175],"underlying":[49],"mechanism?":[50],"After":[51],"series":[53],"of":[54,64,110,177],"preliminary":[55],"experiments":[56],"and":[57,81,113,137,167],"findings,":[58],"suggest":[60],"that":[61,96],"the":[62,74],"risk":[63],"not":[67,141],"caused":[68],"by":[69,73],"length":[70],"itself":[71],"reliance":[76],"on":[77,87,152],"context":[78],"for":[79,107],"coherence":[80],"completeness":[82],"longer":[84,181],"responses.":[85,182],"Building":[86],"these":[88],"insights,":[89],"propose":[91],"novel":[93],"\"induce-detect-suppress\"":[94],"framework":[95,145],"actively":[97],"induces":[98],"through":[100],"deliberately":[101],"designed":[102],"contexts,":[103],"leverages":[104],"induced":[105],"instances":[106],"early":[108],"detection":[109,136],"high-risk":[111],"cases,":[112],"ultimately":[114],"suppresses":[115],"potential":[116],"object-level":[117],"during":[119],"actual":[120],"decoding.":[121],"Our":[122],"approach":[123],"achieves":[124],"consistent,":[125],"improvements":[127],"across":[128],"all":[129],"benchmarks,":[130],"demonstrating":[131],"its":[132],"efficacy.":[133],"The":[134],"strong":[135],"improved":[138],"mitigation":[140],"only":[142],"validate":[143],"our":[144,150],"but,":[146],"importantly,":[148],"re-validate":[149],"hypothesis":[151],"context.":[153],"Rather":[154],"than":[155],"pursuing":[157],"performance":[158],"gains,":[159],"study":[161],"aims":[162],"provide":[164],"new":[165],"insights":[166],"serves":[168],"as":[169],"first":[171],"step":[172],"toward":[173],"exploration":[176],"LVLMs'":[180]},"counts_by_year":[],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-10-25T00:00:00"}