{"id":"https://openalex.org/W4417293464","doi":"https://doi.org/10.1109/wacv61042.2026.00766","title":"SAVE: Sparse Autoencoder-Driven Visual Information Enhancement for Mitigating Object Hallucination","display_name":"SAVE: Sparse Autoencoder-Driven Visual Information Enhancement for Mitigating Object Hallucination","publication_year":2026,"publication_date":"2026-03-06","ids":{"openalex":"https://openalex.org/W4417293464","doi":"https://doi.org/10.1109/wacv61042.2026.00766"},"language":null,"primary_location":{"id":"doi:10.1109/wacv61042.2026.00766","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00766","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2512.07730","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Sangha Park","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Sangha Park","raw_affiliation_strings":["Seoul National University,Department of Electrical and Computer Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University,Department of Electrical and Computer Engineering","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104127570","display_name":"Seungryong Yoo","orcid":null},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seungryong Yoo","raw_affiliation_strings":["Seoul National University,Department of Electrical and Computer Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University,Department of Electrical and Computer Engineering","institution_ids":["https://openalex.org/I139264467"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091880855","display_name":"Jisoo Mok","orcid":"https://orcid.org/0000-0001-7002-0275"},"institutions":[{"id":"https://openalex.org/I193352282","display_name":"Daegu Gyeongbuk Institute of Science and Technology","ror":"https://ror.org/03frjya69","country_code":"KR","type":"education","lineage":["https://openalex.org/I193352282"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jisoo Mok","raw_affiliation_strings":["Daegu Gyeongbuk Institute of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Daegu Gyeongbuk Institute of Science and Technology","institution_ids":["https://openalex.org/I193352282"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086877012","display_name":"Sungroh Yoon","orcid":"https://orcid.org/0000-0002-2367-197X"},"institutions":[{"id":"https://openalex.org/I139264467","display_name":"Seoul National University","ror":"https://ror.org/04h9pn542","country_code":"KR","type":"education","lineage":["https://openalex.org/I139264467"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Sungroh Yoon","raw_affiliation_strings":["Seoul National University,Department of Electrical and Computer Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seoul National University,Department of Electrical and Computer Engineering","institution_ids":["https://openalex.org/I139264467"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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":0.02153463,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7935","last_page":"7944"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.2930000126361847,"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":0.2930000126361847,"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":0.09480000287294388,"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/T11147","display_name":"Misinformation and Its Impacts","score":0.039400000125169754,"subfield":{"id":"https://openalex.org/subfields/3312","display_name":"Sociology and Political Science"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6323000192642212},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.5299000144004822},{"id":"https://openalex.org/keywords/autoencoder","display_name":"Autoencoder","score":0.5281000137329102},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.4803999960422516},{"id":"https://openalex.org/keywords/prior-probability","display_name":"Prior probability","score":0.4596000015735626},{"id":"https://openalex.org/keywords/generalizability-theory","display_name":"Generalizability theory","score":0.44920000433921814},{"id":"https://openalex.org/keywords/visual-hallucination","display_name":"Visual Hallucination","score":0.43320000171661377},{"id":"https://openalex.org/keywords/binary-number","display_name":"Binary number","score":0.35510000586509705},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.3531999886035919}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.66839998960495},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6607000231742859},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6323000192642212},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.5299000144004822},{"id":"https://openalex.org/C101738243","wikidata":"https://www.wikidata.org/wiki/Q786435","display_name":"Autoencoder","level":3,"score":0.5281000137329102},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.4803999960422516},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.4596000015735626},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4512999951839447},{"id":"https://openalex.org/C27158222","wikidata":"https://www.wikidata.org/wiki/Q5532422","display_name":"Generalizability theory","level":2,"score":0.44920000433921814},{"id":"https://openalex.org/C2908998935","wikidata":"https://www.wikidata.org/wiki/Q130741","display_name":"Visual Hallucination","level":2,"score":0.43320000171661377},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.35510000586509705},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3531999886035919},{"id":"https://openalex.org/C3020402766","wikidata":"https://www.wikidata.org/wiki/Q104376712","display_name":"Prior information","level":2,"score":0.33090001344680786},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.3190999925136566},{"id":"https://openalex.org/C160086991","wikidata":"https://www.wikidata.org/wiki/Q5939193","display_name":"Human visual system model","level":3,"score":0.31690001487731934},{"id":"https://openalex.org/C63435697","wikidata":"https://www.wikidata.org/wiki/Q864135","display_name":"Binary code","level":3,"score":0.30169999599456787},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.29670000076293945},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2962000072002411},{"id":"https://openalex.org/C100279451","wikidata":"https://www.wikidata.org/wiki/Q372193","display_name":"Perplexity","level":3,"score":0.27230000495910645},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.2565999925136566},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.25119999051094055},{"id":"https://openalex.org/C2986089797","wikidata":"https://www.wikidata.org/wiki/Q6501338","display_name":"Visual attention","level":3,"score":0.2506999969482422},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.2502000033855438}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/wacv61042.2026.00766","is_oa":false,"landing_page_url":"https://doi.org/10.1109/wacv61042.2026.00766","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2512.07730","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.07730","pdf_url":"https://arxiv.org/pdf/2512.07730","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.2512.07730","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.07730","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:2512.07730","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.07730","pdf_url":"https://arxiv.org/pdf/2512.07730","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":[],"funders":[{"id":"https://openalex.org/F4320320671","display_name":"National Research Foundation","ror":"https://ror.org/05s0g1g46"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Although":[0],"Multimodal":[1],"Large":[2],"Language":[3],"Models":[4],"(MLLMs)":[5],"have":[6],"advanced":[7],"substantially,":[8],"they":[9],"remain":[10],"vulnerable":[11],"to":[12,67,148],"object":[13,143],"hallucination":[14,38],"caused":[15],"by":[16,39],"language":[17],"priors":[18],"and":[19,83,107,112,119,124,145],"visual":[20,63,69,81,135],"information":[21,64],"loss.":[22],"To":[23],"address":[24],"this,":[25],"we":[26],"propose":[27],"SAVE":[28,91],"(Sparse":[29],"Autoencoder-Driven":[30],"Visual":[31],"Information":[32],"Enhancement),":[33],"a":[34,100],"framework":[35],"that":[36,132],"mitigates":[37],"steering":[40,133],"the":[41,55,61,73,122,139],"model":[42,74],"along":[43,75,134],"Sparse":[44],"Autoencoder":[45],"(SAE)":[46],"latent":[47],"features.":[48,71],"A":[49],"binary":[50],"object-presence":[51],"question-answering":[52],"probe":[53],"identifies":[54],"SAE":[56],"features":[57,78,137],"most":[58],"indicative":[59],"of":[60,126,141],"model\u2019s":[62],"processing,":[65],"referred":[66],"as":[68],"understanding":[70,82,136],"Steering":[72],"these":[76],"identified":[77],"reinforces":[79],"grounded":[80],"effectively":[84],"reduces":[85],"hallucination.":[86,152],"With":[87],"its":[88],"simple":[89],"design,":[90],"outperforms":[92],"state-of-the-art":[93],"training-free":[94],"methods":[95],"on":[96,110],"standard":[97],"benchmarks,":[98],"achieving":[99],"10%p":[101],"improvement":[102],"in":[103],"CHAIR<inf":[104],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[105],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">S</inf>":[106],"consistent":[108],"gains":[109],"POPE":[111],"MMHal-Bench.":[113],"Extensive":[114],"evaluations":[115],"across":[116],"multiple":[117],"models":[118],"layers":[120],"confirm":[121],"robustness":[123],"generalizability":[125],"our":[127],"approach.":[128],"Further":[129],"analysis":[130],"reveals":[131],"suppresses":[138],"generation":[140],"uncertain":[142],"tokens":[144],"increases":[146],"attention":[147],"image":[149],"tokens,":[150],"mitigating":[151],"Code":[153],"is":[154],"released":[155],"at":[156],"https://github.com/wiarae/SAVE.":[157]},"counts_by_year":[],"updated_date":"2026-07-03T08:13:44.112507","created_date":"2025-12-10T00:00:00"}