{"id":"https://openalex.org/W4416034282","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.858","title":"MRFD: Multi-Region Fusion Decoding with Self-Consistency for Mitigating Hallucinations in LVLMs","display_name":"MRFD: Multi-Region Fusion Decoding with Self-Consistency for Mitigating Hallucinations in LVLMs","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034282","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.858"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.858","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.858","pdf_url":"https://aclanthology.org/2025.findings-emnlp.858.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.858.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100712413","display_name":"Hong Ge","orcid":"https://orcid.org/0000-0001-9421-2677"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Haonan Ge","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100397304","display_name":"Yiwei Wang","orcid":"https://orcid.org/0000-0001-5921-2575"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yiwei Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100980514","display_name":"Ming-Hsuan Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ming-Hsuan Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5102792044","display_name":"Yujun Cai","orcid":"https://orcid.org/0000-0002-3717-4771"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yujun Cai","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.40101287,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"15860","last_page":"15879"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11797","display_name":"graph theory and CDMA systems","score":0.22190000116825104,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11797","display_name":"graph theory and CDMA systems","score":0.22190000116825104,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10951","display_name":"Cryptographic Implementations and Security","score":0.06120000034570694,"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/T10125","display_name":"Advanced Wireless Communication Techniques","score":0.04050000011920929,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.49709999561309814},{"id":"https://openalex.org/keywords/sensor-fusion","display_name":"Sensor fusion","score":0.3361999988555908},{"id":"https://openalex.org/keywords/fusion","display_name":"Fusion","score":0.319599986076355},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.3151000142097473},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.30799999833106995}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5748000144958496},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.531000018119812},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.49709999561309814},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4154999852180481},{"id":"https://openalex.org/C33954974","wikidata":"https://www.wikidata.org/wiki/Q486494","display_name":"Sensor fusion","level":2,"score":0.3361999988555908},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3325999975204468},{"id":"https://openalex.org/C158525013","wikidata":"https://www.wikidata.org/wiki/Q2593739","display_name":"Fusion","level":2,"score":0.319599986076355},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3151000142097473},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.30799999833106995},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3003000020980835},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.2944999933242798},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.26499998569488525},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.25679999589920044}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.858","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.858","pdf_url":"https://aclanthology.org/2025.findings-emnlp.858.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.858","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.858","pdf_url":"https://aclanthology.org/2025.findings-emnlp.858.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1294669936","display_name":"CRII: RI: Synergistic Learning of Language Models and Knowledge Graphs for Trustworthy Question Answering","funder_award_id":"2451683","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320309480","display_name":"Nvidia","ror":"https://ror.org/03jdj4y14"},{"id":"https://openalex.org/F4320333031","display_name":"University of California Merced","ror":"https://ror.org/00d9ah105"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034282.pdf","grobid_xml":"https://content.openalex.org/works/W4416034282.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Vision-Language":[1],"Models":[2],"(LVLMs)":[3],"have":[4],"shown":[5],"strong":[6],"performance":[7],"across":[8,92],"multimodal":[9],"tasks.However,":[10],"they":[11],"often":[12],"produce":[13],"hallucinations-text":[14],"that":[15,47,98],"is":[16],"inconsistent":[17],"with":[18],"visual":[19],"input,":[20],"due":[21],"to":[22,26],"the":[23,33,75],"limited":[24],"ability":[25],"verify":[27],"information":[28],"in":[29],"different":[30],"regions":[31,57],"of":[32,82],"image.To":[34],"address":[35],"this,":[36],"we":[37],"propose":[38],"Multi-Region":[39],"Fusion":[40],"Decoding":[41],"(MRFD),":[42],"a":[43,79],"trainingfree":[44],"decoding":[45],"method":[46],"improves":[48,104],"factual":[49],"grounding":[50],"by":[51,89],"modeling":[52],"inter-region":[53],"consistency.MRFD":[54],"identifies":[55],"salient":[56],"using":[58,85],"cross-attention,":[59],"generates":[60],"initial":[61],"responses":[62],"for":[63],"each,":[64],"and":[65,95,103],"computes":[66],"reliability":[67],"weights":[68,77],"based":[69],"on":[70],"Jensen-Shannon":[71],"Divergence":[72],"(JSD)":[73],"among":[74],"responses.These":[76],"guide":[78],"consistencyaware":[80],"fusion":[81],"per-region":[83],"predictions,":[84],"region-aware":[86],"prompts":[87],"inspired":[88],"Chain-of-Thought":[90],"reasoning.Experiments":[91],"multiple":[93],"LVLMs":[94],"benchmarks":[96],"show":[97],"MRFD":[99],"significantly":[100],"reduces":[101],"hallucinations":[102],"response":[105],"factuality":[106],"without":[107],"requiring":[108],"model":[109],"updates.":[110]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-11-08T00:00:00"}