{"id":"https://openalex.org/W7166840263","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1241","title":"Look and Think: Efficient Multimodal Reasoning via Modality-Decoupled Compression","display_name":"Look and Think: Efficient Multimodal Reasoning via Modality-Decoupled Compression","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166840263","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1241"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1241","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1241","pdf_url":"https://aclanthology.org/2026.findings-acl.1241.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.1241.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5121927805","display_name":"Xidi Cai","orcid":null},"institutions":[{"id":"https://openalex.org/I90610280","display_name":"South China University of Technology","ror":"https://ror.org/0530pts50","country_code":"CN","type":"education","lineage":["https://openalex.org/I90610280"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xidi Cai","raw_affiliation_strings":["School of Computer Science and Engineering , South China University of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering , South China University of Technology","institution_ids":["https://openalex.org/I90610280"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139720161","display_name":"Junhao Zheng","orcid":null},"institutions":[{"id":"https://openalex.org/I90610280","display_name":"South China University of Technology","ror":"https://ror.org/0530pts50","country_code":"CN","type":"education","lineage":["https://openalex.org/I90610280"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Junhao Zheng","raw_affiliation_strings":["School of Computer Science and Engineering , South China University of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering , South China University of Technology","institution_ids":["https://openalex.org/I90610280"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114833340","display_name":"Jingye Li","orcid":"https://orcid.org/0000-0002-9063-3594"},"institutions":[{"id":"https://openalex.org/I90610280","display_name":"South China University of Technology","ror":"https://ror.org/0530pts50","country_code":"CN","type":"education","lineage":["https://openalex.org/I90610280"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jingye Li","raw_affiliation_strings":["School of Computer Science and Engineering , South China University of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering , South China University of Technology","institution_ids":["https://openalex.org/I90610280"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139791395","display_name":"Boyuan Li","orcid":null},"institutions":[{"id":"https://openalex.org/I90610280","display_name":"South China University of Technology","ror":"https://ror.org/0530pts50","country_code":"CN","type":"education","lineage":["https://openalex.org/I90610280"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Boyuan Li","raw_affiliation_strings":["School of Computer Science and Engineering , South China University of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering , South China University of Technology","institution_ids":["https://openalex.org/I90610280"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139721623","display_name":"Shaowei Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I90610280","display_name":"South China University of Technology","ror":"https://ror.org/0530pts50","country_code":"CN","type":"education","lineage":["https://openalex.org/I90610280"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shaowei Zhang","raw_affiliation_strings":["School of Computer Science and Engineering , South China University of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering , South China University of Technology","institution_ids":["https://openalex.org/I90610280"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139797922","display_name":"Qianli Ma","orcid":null},"institutions":[{"id":"https://openalex.org/I90610280","display_name":"South China University of Technology","ror":"https://ror.org/0530pts50","country_code":"CN","type":"education","lineage":["https://openalex.org/I90610280"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qianli Ma","raw_affiliation_strings":["School of Computer Science and Engineering , South China University of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering , South China University of Technology","institution_ids":["https://openalex.org/I90610280"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I90610280"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.82968933,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"24787","last_page":"24806"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.33169999718666077,"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/T12031","display_name":"Speech and dialogue systems","score":0.33169999718666077,"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/T11596","display_name":"Constraint Satisfaction and Optimization","score":0.19939999282360077,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.15209999680519104,"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/compression","display_name":"Compression (physics)","score":0.4977000057697296},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.4348999857902527},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.24729999899864197},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.24639999866485596},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.24009999632835388}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6353999972343445},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.4977000057697296},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.482699990272522},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.4348999857902527},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.31349998712539673},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2870999872684479},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.24729999899864197},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.24639999866485596},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.24009999632835388},{"id":"https://openalex.org/C57654395","wikidata":"https://www.wikidata.org/wiki/Q1097775","display_name":"Compression artifact","level":5,"score":0.22179999947547913}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1241","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1241","pdf_url":"https://aclanthology.org/2026.findings-acl.1241.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.1241","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1241","pdf_url":"https://aclanthology.org/2026.findings-acl.1241.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":[{"id":"https://openalex.org/G1087259888","display_name":null,"funder_award_id":"2023YFA1011601","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G1656231566","display_name":null,"funder_award_id":"62273109","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G1800336020","display_name":null,"funder_award_id":"62272173","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G5663721638","display_name":null,"funder_award_id":"2023A0505050106","funder_id":"https://openalex.org/F4320335795","funder_display_name":"Science and Technology Planning Project of Guangdong Province"},{"id":"https://openalex.org/G5967704285","display_name":null,"funder_award_id":"2024A1515010089","funder_id":"https://openalex.org/F4320321921","funder_display_name":"Natural Science Foundation of Guangdong Province"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320321921","display_name":"Natural Science Foundation of Guangdong Province","ror":null},{"id":"https://openalex.org/F4320335795","display_name":"Science and Technology Planning Project of Guangdong Province","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166840263.pdf","grobid_xml":"https://content.openalex.org/works/W7166840263.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0],"large":[1,34],"language":[2],"models":[3],"(MLLMs)":[4],"have":[5],"achieved":[6],"strong":[7],"performance":[8,133],"on":[9,88],"challenging":[10],"visual":[11,37,84,94,97],"question":[12],"answering":[13],"benchmarks,":[14],"yet":[15],"their":[16],"inference":[17],"efficiency":[18],"is":[19,86,99],"severely":[20],"constrained":[21],"by":[22,127],"the":[23,33,44,62,111,123,136],"rapidly":[24],"growing":[25],"context.This":[26],"growth":[27,116],"stems":[28],"from":[29],"two":[30,112],"primary":[31],"sources:":[32],"number":[35],"of":[36,46,114],"tokens":[38,95],"required":[39],"to":[40,129,135],"encode":[41],"images,":[42],"and":[43,60,77,101],"accumulation":[45],"intermediate":[47],"reasoning":[48,73],"traces":[49],"during":[50],"autoregressive":[51],"generation.To":[52],"address":[53],"these":[54],"challenges,":[55],"we":[56],"propose":[57],"LaT":[58,91,121],"(Look":[59],"Think),":[61],"first":[63],"modality-decoupled":[64],"compression":[65,105],"method":[66],"that":[67,120],"enables":[68],"efficient":[69],"multimodal":[70],"inference.LaT":[71],"structures":[72],"into":[74],"alternating":[75],"looking":[76],"thinking":[78],"steps,":[79],"thereby":[80],"explicitly":[81],"signaling":[82],"when":[83],"grounding":[85,98],"required.Building":[87],"this":[89],"design,":[90],"(1)":[92],"evicts":[93],"whenever":[96],"unnecessary,":[100],"(2)":[102],"applies":[103],"co-learning-guided":[104],"after":[106],"each":[107],"completed":[108],"step,":[109],"mitigating":[110],"sources":[113],"context":[115,125],"respectively.Experimental":[117],"results":[118],"demonstrate":[119],"reduces":[122],"average":[124],"length":[126],"up":[128],"57%,":[130],"while":[131],"maintaining":[132],"comparable":[134],"standard":[137],"MLLM.":[138]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}
