{"id":"https://openalex.org/W7214927062","doi":"https://doi.org/10.48550/arxiv.2609.37263","title":"Beyond Attention Imbalance: Mitigating Hallucinations via Spectral Surgery","display_name":"Beyond Attention Imbalance: Mitigating Hallucinations via Spectral Surgery","publication_year":2026,"publication_date":"2026-09-29","ids":{"openalex":"https://openalex.org/W7214927062","doi":"https://doi.org/10.48550/arxiv.2609.37263"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.37263","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.37263","pdf_url":"https://arxiv.org/pdf/2609.37263","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,"listed_in":[],"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"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.37263","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5143950757","display_name":"Siqi Lu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Siqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153627886","display_name":"Suo Wei","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Suo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5046296571","display_name":"Yongbin Zheng","orcid":"https://orcid.org/0000-0002-3631-7929","observed_orcids":["https://orcid.org/0000-0002-3631-7929"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Yongbin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125904303","display_name":"Jianhang Yao","orcid":"https://orcid.org/0009-0003-0044-8315","observed_orcids":["https://orcid.org/0009-0003-0044-8315"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yao, Jianhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100912936","display_name":"Wanying Xu","orcid":"https://orcid.org/0000-0003-3698-506X","observed_orcids":["https://orcid.org/0000-0003-3698-506X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Wanying","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5153595130","display_name":"Peng Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Peng","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":true,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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/T11605","display_name":"Visual Attention and Saliency Detection","score":0.0,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/hallucination-mitigation","display_name":"hallucination mitigation","score":0.9649999737739563},{"id":"https://openalex.org/keywords/large-vision-language-models","display_name":"large vision-language models","score":0.9869999885559082},{"id":"https://openalex.org/keywords/spectral-modulation","display_name":"spectral modulation","score":0.7149999737739563},{"id":"https://openalex.org/keywords/contrastive-decoding","display_name":"contrastive decoding","score":0.6039999723434448}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5374000072479248},{"id":"https://openalex.org/C102931765","wikidata":"https://www.wikidata.org/wiki/Q189673","display_name":"Dissociation (chemistry)","level":2,"score":0.48669999837875366},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.45509999990463257},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.45249998569488525},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44359999895095825},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.4253000020980835},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.39959999918937683},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3497999906539917},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.34470000863075256},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.32710000872612},{"id":"https://openalex.org/C2908998935","wikidata":"https://www.wikidata.org/wiki/Q130741","display_name":"Visual Hallucination","level":2,"score":0.3181000053882599},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.31779998540878296},{"id":"https://openalex.org/C2777526259","wikidata":"https://www.wikidata.org/wiki/Q221836","display_name":"Flash (photography)","level":2,"score":0.3127000033855438},{"id":"https://openalex.org/C123079801","wikidata":"https://www.wikidata.org/wiki/Q750240","display_name":"Modulation (music)","level":2,"score":0.3000999987125397},{"id":"https://openalex.org/C2986089797","wikidata":"https://www.wikidata.org/wiki/Q6501338","display_name":"Visual attention","level":3,"score":0.2815000116825104},{"id":"https://openalex.org/C2779918689","wikidata":"https://www.wikidata.org/wiki/Q3771842","display_name":"Stimulus (psychology)","level":2,"score":0.2802000045776367},{"id":"https://openalex.org/C2983668108","wikidata":"https://www.wikidata.org/wiki/Q280453","display_name":"Spectral analysis","level":3,"score":0.27480000257492065},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.27399998903274536},{"id":"https://openalex.org/C56461940","wikidata":"https://www.wikidata.org/wiki/Q970687","display_name":"Eye tracking","level":2,"score":0.26440000534057617},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.2603999972343445},{"id":"https://openalex.org/C2779136372","wikidata":"https://www.wikidata.org/wiki/Q10283002","display_name":"Information flow","level":2,"score":0.25049999356269836}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.37263","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.37263","pdf_url":"https://arxiv.org/pdf/2609.37263","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,"listed_in":[],"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.2609.37263","doi":"https://doi.org/10.48550/arxiv.2609.37263","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.37263","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.37263","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.37263","pdf_url":"https://arxiv.org/pdf/2609.37263","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,"listed_in":[],"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":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G4829830226","display_name":null,"funder_award_id":"62273353","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":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214927062.pdf","grobid_xml":"https://content.openalex.org/works/W7214927062.grobid-xml"},"referenced_works_count":31,"referenced_works":["https://openalex.org/W3034175346","https://openalex.org/W3034771037","https://openalex.org/W3213192039","https://openalex.org/W4312846625","https://openalex.org/W4366850747","https://openalex.org/W4382142077","https://openalex.org/W4386555477","https://openalex.org/W4386875326","https://openalex.org/W4387113775","https://openalex.org/W4388481390","https://openalex.org/W4388685586","https://openalex.org/W4389157024","https://openalex.org/W4389983382","https://openalex.org/W4393160204","https://openalex.org/W4402670231","https://openalex.org/W4402727405","https://openalex.org/W4402727764","https://openalex.org/W4402727851","https://openalex.org/W4404575065","https://openalex.org/W4404782590","https://openalex.org/W4404783798","https://openalex.org/W4405596328","https://openalex.org/W4406490324","https://openalex.org/W4407194361","https://openalex.org/W4414198367","https://openalex.org/W4414349124","https://openalex.org/W4415085973","https://openalex.org/W4415966298","https://openalex.org/W6929419839","https://openalex.org/W6929454498","https://openalex.org/W7133208539"],"related_works":["https://openalex.org/W7155417333","https://openalex.org/W4415112052","https://openalex.org/W4405626860","https://openalex.org/W6892055410","https://openalex.org/W7162606907","https://openalex.org/W7155089030","https://openalex.org/W4417447638","https://openalex.org/W4404260582","https://openalex.org/W4416034529","https://openalex.org/W4407123999"],"abstract_inverted_index":{"While":[0],"Large":[1],"Vision-Language":[2],"Models":[3],"(LVLMs)":[4],"achieve":[5],"remarkable":[6],"success,":[7],"hallucinations":[8],"remain":[9],"a":[10,83,90,121],"significant":[11],"barrier":[12],"to":[13,23,94,107,129],"their":[14],"reliable":[15],"deployment.":[16],"Recent":[17],"studies":[18],"primarily":[19],"attribute":[20],"these":[21],"issues":[22],"cross-modal":[24],"attention":[25,100],"imbalances;":[26],"most":[27],"solutions":[28],"therefore":[29],"focus":[30],"on":[31,55],"reweighting":[32],"visual":[33,49,110],"tokens":[34],"or":[35],"suppressing":[36],"language":[37],"priors.":[38],"However,":[39],"such":[40],"approaches":[41],"often":[42],"overlook":[43],"the":[44,48,109],"spectral":[45,105],"characteristics":[46],"of":[47,64],"information":[50,111],"flow":[51,112],"and":[52,75,85,102,126],"frequently":[53],"rely":[54],"Contrastive":[56],"Decoding":[57],"(CD),":[58],"which":[59],"doubles":[60],"inference":[61],"time.":[62],"Instead":[63],"following":[65],"conventional":[66],"approaches,":[67],"we":[68],"identify":[69],"two":[70],"distinct":[71],"hallucination":[72],"patterns-Perceptual-Semantic":[73],"Dissociation":[74],"Localized":[76],"Fixation-and":[77],"propose":[78],"FLASH":[79,88,119],"(Frequency-Localized":[80],"Attention":[81],"SHaping),":[82],"training-free":[84],"CD-free":[86],"framework.":[87],"utilizes":[89],"Spectral":[91],"Vortex":[92],"Score":[93],"detect":[95],"vision":[96],"heads":[97],"within":[98],"multi-head":[99],"layers":[101],"applies":[103],"adaptive":[104],"modulation":[106],"rectify":[108],"during":[113],"decoding.":[114],"Empirical":[115],"results":[116],"demonstrate":[117],"that":[118],"achieves":[120],"superior":[122],"balance":[123],"between":[124],"performance":[125],"efficiency":[127],"compared":[128],"SOTA":[130],"methods.":[131]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-01T00:00:00"}