{"id":"https://openalex.org/W7166862086","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1253","title":"Token-level Inference-Time Alignment for Vision-Language Models","display_name":"Token-level Inference-Time Alignment for Vision-Language Models","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166862086","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1253"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1253","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1253","pdf_url":"https://aclanthology.org/2026.findings-acl.1253.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: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.1253.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5123525275","display_name":"Kejia Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I4210156488","display_name":"Institute of Zoology of the Slovak Academy of Sciences","ror":"https://ror.org/05tm6j853","country_code":"SK","type":"facility","lineage":["https://openalex.org/I207624831","https://openalex.org/I4210156488"]},{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN","SK"],"is_corresponding":false,"raw_author_name":"Kejia Chen","raw_affiliation_strings":["Hangzhou High-Tech Zone (Binjiang ) Institute of Blockchain and Data Security","The State Key Laboratory of Blockchain and Data Security , Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hangzhou High-Tech Zone (Binjiang ) Institute of Blockchain and Data Security","institution_ids":["https://openalex.org/I4210156488"]},{"raw_affiliation_string":"The State Key Laboratory of Blockchain and Data Security , Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102516938","display_name":"Junjun Zheng","orcid":null},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Junjun Zheng","raw_affiliation_strings":["Taobao & Tmall Group , Alibaba Inc. 4 Sun Yat-sen"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Taobao & Tmall Group , Alibaba Inc. 4 Sun Yat-sen","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139773048","display_name":"Jiawen Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210156488","display_name":"Institute of Zoology of the Slovak Academy of Sciences","ror":"https://ror.org/05tm6j853","country_code":"SK","type":"facility","lineage":["https://openalex.org/I207624831","https://openalex.org/I4210156488"]},{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN","SK"],"is_corresponding":false,"raw_author_name":"Jiawen Zhang","raw_affiliation_strings":["Hangzhou High-Tech Zone (Binjiang ) Institute of Blockchain and Data Security","The State Key Laboratory of Blockchain and Data Security , Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hangzhou High-Tech Zone (Binjiang ) Institute of Blockchain and Data Security","institution_ids":["https://openalex.org/I4210156488"]},{"raw_affiliation_string":"The State Key Laboratory of Blockchain and Data Security , Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015803667","display_name":"Manxi Lin","orcid":"https://orcid.org/0000-0003-3399-8682"},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Manxi Lin","raw_affiliation_strings":["Taobao & Tmall Group , Alibaba Inc. 4 Sun Yat-sen"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Taobao & Tmall Group , Alibaba Inc. 4 Sun Yat-sen","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013919453","display_name":"Xiao Pan","orcid":"https://orcid.org/0000-0001-9873-5520"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao Pan","raw_affiliation_strings":["University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139811675","display_name":"Jiacong Hu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210156488","display_name":"Institute of Zoology of the Slovak Academy of Sciences","ror":"https://ror.org/05tm6j853","country_code":"SK","type":"facility","lineage":["https://openalex.org/I207624831","https://openalex.org/I4210156488"]},{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN","SK"],"is_corresponding":false,"raw_author_name":"Jiacong Hu","raw_affiliation_strings":["Hangzhou High-Tech Zone (Binjiang ) Institute of Blockchain and Data Security","The State Key Laboratory of Blockchain and Data Security , Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hangzhou High-Tech Zone (Binjiang ) Institute of Blockchain and Data Security","institution_ids":["https://openalex.org/I4210156488"]},{"raw_affiliation_string":"The State Key Laboratory of Blockchain and Data Security , Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139757346","display_name":"Jian Lou","orcid":null},"institutions":[{"id":"https://openalex.org/I180726961","display_name":"Shenzhen University","ror":"https://ror.org/01vy4gh70","country_code":"CN","type":"education","lineage":["https://openalex.org/I180726961"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jian Lou","raw_affiliation_strings":["Shenzhen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shenzhen University","institution_ids":["https://openalex.org/I180726961"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139740051","display_name":"Zunlei Feng","orcid":null},"institutions":[{"id":"https://openalex.org/I4210156488","display_name":"Institute of Zoology of the Slovak Academy of Sciences","ror":"https://ror.org/05tm6j853","country_code":"SK","type":"facility","lineage":["https://openalex.org/I207624831","https://openalex.org/I4210156488"]},{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN","SK"],"is_corresponding":false,"raw_author_name":"Zunlei Feng","raw_affiliation_strings":["Hangzhou High-Tech Zone (Binjiang ) Institute of Blockchain and Data Security","The State Key Laboratory of Blockchain and Data Security , Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hangzhou High-Tech Zone (Binjiang ) Institute of Blockchain and Data Security","institution_ids":["https://openalex.org/I4210156488"]},{"raw_affiliation_string":"The State Key Laboratory of Blockchain and Data Security , Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139721274","display_name":"Mingli Song","orcid":null},"institutions":[{"id":"https://openalex.org/I4210156488","display_name":"Institute of Zoology of the Slovak Academy of Sciences","ror":"https://ror.org/05tm6j853","country_code":"SK","type":"facility","lineage":["https://openalex.org/I207624831","https://openalex.org/I4210156488"]},{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN","SK"],"is_corresponding":false,"raw_author_name":"Mingli Song","raw_affiliation_strings":["Hangzhou High-Tech Zone (Binjiang ) Institute of Blockchain and Data Security","The State Key Laboratory of Blockchain and Data Security , Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hangzhou High-Tech Zone (Binjiang ) Institute of Blockchain and Data Security","institution_ids":["https://openalex.org/I4210156488"]},{"raw_affiliation_string":"The State Key Laboratory of Blockchain and Data Security , Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":4,"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":0.85085998,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"25012","last_page":"25029"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.5784000158309937,"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":0.5784000158309937,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.08839999884366989,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.05790000036358833,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.2791000008583069},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.24799999594688416},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.24719999730587006},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.21819999814033508}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4878000020980835},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.35370001196861267},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.2791000008583069},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2551000118255615},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.24799999594688416},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.24719999730587006},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.24400000274181366},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.21819999814033508},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.21799999475479126},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.21789999306201935}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1253","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1253","pdf_url":"https://aclanthology.org/2026.findings-acl.1253.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: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.1253","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1253","pdf_url":"https://aclanthology.org/2026.findings-acl.1253.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: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166862086.pdf","grobid_xml":"https://content.openalex.org/works/W7166862086.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Vision-Language":[0],"Models":[1],"(VLMs)":[2],"often":[3],"prioritize":[4],"linguistic":[5],"fluency":[6],"over":[7],"visual":[8,69],"fidelity,":[9],"leading":[10],"to":[11,67,92],"hallucinations":[12],"where":[13],"generated":[14],"text":[15],"contradicts":[16],"the":[17,34,53,58,94],"image.Countering":[18],"this":[19],"bias":[20],"typically":[21],"requires":[22],"resource-heavy":[23],"fine-tuning":[24],"or":[25],"high-latency":[26],"verification":[27],"methods":[28],"that":[29,51],"provide":[30],"feedback":[31,91],"only":[32],"after":[33],"full":[35],"response":[36],"is":[37,149],"generated.To":[38],"overcome":[39],"these":[40,130],"limitations,":[41],"we":[42],"present":[43],"a":[44,63,124,140],"framework":[45],"for":[46],"Token-level":[47],"Inference-Time":[48],"Alignment":[49],"(TITA)":[50],"steers":[52],"decoding":[54],"process":[55],"without":[56],"updating":[57],"base":[59],"model":[60,66],"parameters.By":[61],"training":[62],"lightweight":[64],"reward":[65],"capture":[68],"preferences,":[70],"TITA":[71,115],"extracts":[72],"implicit":[73],"guidance":[74],"through":[75],"log-probability":[76],"ratios.This":[77],"approach":[78],"functions":[79],"as":[80],"an":[81],"inferencetime":[82],"adaptation":[83],"of":[84],"Direct":[85],"Preference":[86],"Optimization":[87],"(DPO),":[88],"injecting":[89],"dense":[90],"correct":[93],"output":[95],"distribution":[96],"at":[97],"every":[98],"generation":[99],"step.Across":[100],"diverse":[101],"architectures":[102],"including":[103],"LLaVA-1.5,":[104],"Qwen3-VL,":[105],"and":[106,122,146],"InternVL3.5,TITA":[107],"consistently":[108],"improves":[109],"performance":[110],"on":[111,120],"13":[112],"benchmarks.For":[113],"example,":[114],"boosts":[116],"LLaVA-1.5-7B":[117],"by":[118],"+8.6%":[119],"MMVet":[121],"achieves":[123],"74.0":[125],"MMStar":[126],"score":[127],"with":[128],"Qwen3-VL-8B.Specifically,":[129],"gains":[131],"incur":[132],"negligible":[133],"overhead":[134],"(":[135],"0.2s":[136],"per":[137],"query),":[138],"offering":[139],"superior":[141],"tradeoff":[142],"between":[143],"alignment":[144],"effectiveness":[145],"efficiency.Our":[147],"code":[148],"available":[150],"at:":[151],"https:":[152],"//github.com/Thecommonirin/TITA":[153]},"counts_by_year":[],"updated_date":"2026-08-22T07:34:49.880490","created_date":"2026-07-02T00:00:00"}
