{"id":"https://openalex.org/W7161296481","doi":"https://doi.org/10.48550/arxiv.2605.14966","title":"MHSA: A Lightweight Framework for Mitigating Hallucinations via Steered Attention in LVLMs","display_name":"MHSA: A Lightweight Framework for Mitigating Hallucinations via Steered Attention in LVLMs","publication_year":2026,"publication_date":"2026-05-14","ids":{"openalex":"https://openalex.org/W7161296481","doi":"https://doi.org/10.48550/arxiv.2605.14966"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.14966","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14966","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.2605.14966","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5088432110","display_name":"Wei Ding","orcid":"https://orcid.org/0000-0002-3383-551X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ding, Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136234682","display_name":"Yilin Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yilin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136227460","display_name":"Yudong Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yudong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101577090","display_name":"Ruobing Xie","orcid":"https://orcid.org/0000-0003-3170-5647"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Ruobing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039337290","display_name":"Xingwu Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Xingwu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136202445","display_name":"Jiansheng Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Jiansheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136261709","display_name":"Yu Wang","orcid":"https://orcid.org/0009-0004-4838-7432"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yu","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/T13397","display_name":"Hallucinations in medical conditions","score":0.09229999780654907,"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"}},"topics":[{"id":"https://openalex.org/T13397","display_name":"Hallucinations in medical conditions","score":0.09229999780654907,"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"}},{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.06790000200271606,"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/T10023","display_name":"Schizophrenia research and treatment","score":0.066600002348423,"subfield":{"id":"https://openalex.org/subfields/2738","display_name":"Psychiatry and Mental health"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/visual-hallucination","display_name":"Visual Hallucination","score":0.6506999731063843},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.6100999712944031},{"id":"https://openalex.org/keywords/perspective","display_name":"Perspective (graphical)","score":0.5906999707221985},{"id":"https://openalex.org/keywords/discriminator","display_name":"Discriminator","score":0.49309998750686646},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.39259999990463257},{"id":"https://openalex.org/keywords/generator","display_name":"Generator (circuit theory)","score":0.3862999975681305}],"concepts":[{"id":"https://openalex.org/C2908998935","wikidata":"https://www.wikidata.org/wiki/Q130741","display_name":"Visual Hallucination","level":2,"score":0.6506999731063843},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.6100999712944031},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.5906999707221985},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5633999705314636},{"id":"https://openalex.org/C2779803651","wikidata":"https://www.wikidata.org/wiki/Q5282088","display_name":"Discriminator","level":3,"score":0.49309998750686646},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.39259999990463257},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.3862999975681305},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.36970001459121704},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.367000013589859},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.35740000009536743},{"id":"https://openalex.org/C26876914","wikidata":"https://www.wikidata.org/wiki/Q11166","display_name":"Dynamic Host Configuration Protocol","level":3,"score":0.3481999933719635},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3248000144958496},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2621000111103058},{"id":"https://openalex.org/C169900460","wikidata":"https://www.wikidata.org/wiki/Q2200417","display_name":"Cognition","level":2,"score":0.2603999972343445},{"id":"https://openalex.org/C101738243","wikidata":"https://www.wikidata.org/wiki/Q786435","display_name":"Autoencoder","level":3,"score":0.25589999556541443}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.14966","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14966","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.2605.14966","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14966","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":[{"display_name":"Reduced inequalities","score":0.7536574006080627,"id":"https://metadata.un.org/sdg/10"}],"awards":[],"funders":[],"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],"achieved":[5],"remarkable":[6],"performance":[7],"across":[8,111],"diverse":[9],"multimodal":[10],"tasks,":[11],"yet":[12],"they":[13],"continue":[14],"to":[15,72,86,140],"suffer":[16],"from":[17,41,94,137],"hallucinations,":[18],"generating":[19],"content":[20],"that":[21,67],"is":[22],"inconsistent":[23],"with":[24,123],"the":[25,42,95,99,119,124],"visual":[26],"input.":[27],"Prior":[28],"work":[29],"DHCP":[30,96],"(Detecting":[31],"Hallucinations":[32,60],"by":[33,70,91,116],"Cross-modal":[34],"Attention":[35],"Pattern)":[36],"has":[37],"explored":[38],"hallucination":[39,51,138,141,149],"detection":[40,139],"perspective":[43,147],"of":[44],"cross-modal":[45,74,121,134],"attention,":[46,89],"but":[47],"does":[48],"not":[49],"address":[50],"mitigation.":[52],"In":[53],"this":[54],"paper,":[55],"we":[56],"propose":[57],"MHSA":[58,79,104,143],"(Mitigating":[59],"via":[61],"Steered":[62],"Attention),":[63],"a":[64,81,145],"lightweight":[65],"framework":[66],"mitigates":[68,105],"hallucinations":[69,110],"learning":[71],"correct":[73],"attention":[75,122,135],"patterns":[76],"in":[77,151],"LVLMs.":[78],"trains":[80],"simple":[82],"three-layer":[83],"MLP":[84],"generator":[85],"produce":[87],"corrected":[88,125],"guided":[90],"supervisory":[92],"signals":[93],"discriminator":[97],"and":[98,108,114,153],"LVLM":[100,130],"itself.":[101],"During":[102],"inference,":[103],"both":[106],"discriminative":[107],"generative":[109],"various":[112],"datasets":[113],"LVLMs":[115,152],"simply":[117],"replacing":[118],"original":[120],"one,":[126],"without":[127],"modifying":[128],"any":[129],"parameters.":[131],"By":[132],"extending":[133],"mechanisms":[136],"mitigation,":[142],"offers":[144],"novel":[146],"on":[148],"research":[150],"helps":[154],"enhance":[155],"their":[156],"reliability.":[157]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-05-16T00:00:00"}
