{"id":"https://openalex.org/W4417302315","doi":"https://doi.org/10.18653/v1/2026.findings-acl.223","title":"TinyAlign: Boosting Lightweight Vision-Language Models by Mitigating Modal Alignment Bottlenecks","display_name":"TinyAlign: Boosting Lightweight Vision-Language Models by Mitigating Modal Alignment Bottlenecks","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4417302315","doi":"https://doi.org/10.18653/v1/2026.findings-acl.223"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.223","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.223","pdf_url":"https://aclanthology.org/2026.findings-acl.223.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":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.223.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5049986768","display_name":"Yifan Hu","orcid":"https://orcid.org/0000-0002-9332-1852"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuanze Hu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5143239705","display_name":"Xinyu Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xinyu Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101448608","display_name":"Zhichao Yang","orcid":"https://orcid.org/0009-0008-3398-1286"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhichao Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100733093","display_name":"Siwei Li","orcid":"https://orcid.org/0000-0003-1985-8149"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gen Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102004189","display_name":"Ye Qiu","orcid":"https://orcid.org/0000-0002-2306-7342"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ye Qiu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113151537","display_name":"Zhaoxin Fan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhaoxin Fan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5143350654","display_name":"Yifan Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yifan Sun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064640873","display_name":"Wenjun Wu","orcid":"https://orcid.org/0000-0001-9594-7026"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wenjun wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5143212583","display_name":"Jin Dong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jin Dong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100638710","display_name":"Xiaotie Deng","orcid":"https://orcid.org/0000-0002-5282-6467"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaotie Deng","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":0.03236777,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4585","last_page":"4597"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.7730000019073486,"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":0.7730000019073486,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.06970000267028809,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.04149999842047691,"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/boosting","display_name":"Boosting (machine learning)","score":0.7804999947547913},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.703499972820282},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.526199996471405},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4997999966144562},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.4293999969959259},{"id":"https://openalex.org/keywords/through-the-lens-metering","display_name":"Through-the-lens metering","score":0.4242999851703644},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.39559999108314514},{"id":"https://openalex.org/keywords/modal","display_name":"Modal","score":0.38100001215934753}],"concepts":[{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.7804999947547913},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7698000073432922},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.703499972820282},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.526199996471405},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4997999966144562},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47600001096725464},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.4293999969959259},{"id":"https://openalex.org/C43091099","wikidata":"https://www.wikidata.org/wiki/Q1067788","display_name":"Through-the-lens metering","level":3,"score":0.4242999851703644},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.39559999108314514},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38179999589920044},{"id":"https://openalex.org/C71139939","wikidata":"https://www.wikidata.org/wiki/Q910194","display_name":"Modal","level":2,"score":0.38100001215934753},{"id":"https://openalex.org/C2983448237","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Language understanding","level":2,"score":0.3792000114917755},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.31299999356269836},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3125},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.3010999858379364},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.28459998965263367},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2824999988079071},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2680000066757202},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.26460000872612},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.2533999979496002},{"id":"https://openalex.org/C152139883","wikidata":"https://www.wikidata.org/wiki/Q252973","display_name":"Mutual information","level":2,"score":0.2531000077724457}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.223","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.223","pdf_url":"https://aclanthology.org/2026.findings-acl.223.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"},{"id":"pmh:oai:arXiv.org:2505.12884","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.12884","pdf_url":"https://arxiv.org/pdf/2505.12884","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2505.12884","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2505.12884","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:10.18653/v1/2026.findings-acl.223","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.223","pdf_url":"https://aclanthology.org/2026.findings-acl.223.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":[{"id":"https://openalex.org/F4320322499","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4417302315.pdf","grobid_xml":"https://content.openalex.org/works/W4417302315.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Lightweight":[0],"Vision-Language":[1],"Models":[2],"(VLMs)":[3],"are":[4],"indispensable":[5],"for":[6,45,69,151],"resource-constrained":[7],"applications.The":[8],"prevailing":[9],"lightweight":[10,78,153],"recipe":[11,150],"freezes":[12],"both":[13,147],"the":[14,18,33,38,70,142],"vision":[15],"encoder":[16],"and":[17,41,59,105,123,155],"language":[19,39,79],"model":[20,40,80],"while":[21],"training":[22,103,119],"only":[23,139],"a":[24,66,76,90,98,148],"small":[25],"connector.While":[26],"efficient,":[27],"this":[28,48,52,85],"strategy":[29],"relies":[30],"heavily":[31],"on":[32,159],"intrinsic":[34],"representational":[35],"capacity":[36],"of":[37,72,141],"can":[42,81],"become":[43],"suboptimal":[44],"compact":[46],"backbones.In":[47],"work,":[49],"we":[50,87],"study":[51],"alignment":[53,160],"bottleneck":[54],"through":[55],"an":[56,156],"information-theoretic":[57],"lens":[58],"use":[60],"Effective":[61],"Mutual":[62],"Information":[63],"(EMI)":[64],"as":[65],"guiding":[67],"intuition":[68],"amount":[71],"multimodal":[73,108,164],"information":[74],"that":[75,93,115],"frozen":[77],"effectively":[82],"exploit.Motivated":[83],"by":[84],"perspective,":[86],"propose":[88],"TinyAlign,":[89],"retrieval-augmented":[91],"framework":[92],"retrieves":[94],"relevant":[95],"context":[96],"from":[97,102],"memory":[99],"bank":[100],"constructed":[101],"data":[104],"injects":[106],"compressed":[107],"cues":[109],"to":[110],"ease":[111],"alignment.Extensive":[112],"experiments":[113],"show":[114],"TinyAlign":[116],"consistently":[117],"reduces":[118],"loss,":[120],"accelerates":[121],"convergence,":[122],"improves":[124],"downstream":[125],"performance":[126,137],"with":[127,138],"negligible":[128],"computational":[129],"overhead.TinyAlign":[130],"is":[131],"also":[132],"highly":[133],"dataefficient,":[134],"reaching":[135],"baseline-level":[136],"40%":[140],"fine-tuning":[143],"data.These":[144],"results":[145],"provide":[146],"practical":[149],"stronger":[152],"VLMs":[154],"informative":[157],"perspective":[158],"bottlenecks":[161],"in":[162],"constrained":[163],"systems.":[165]},"counts_by_year":[],"updated_date":"2026-08-05T07:39:15.569665","created_date":"2025-10-10T00:00:00"}