{"id":"https://openalex.org/W7133303058","doi":"https://doi.org/10.48550/arxiv.2603.01490","title":"ATA: Bridging Implicit Reasoning with Attention-Guided and Action-Guided Inference for Vision-Language Action Models","display_name":"ATA: Bridging Implicit Reasoning with Attention-Guided and Action-Guided Inference for Vision-Language Action Models","publication_year":2026,"publication_date":"2026-03-02","ids":{"openalex":"https://openalex.org/W7133303058","doi":"https://doi.org/10.48550/arxiv.2603.01490"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.01490","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.01490","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":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.01490","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5127946184","display_name":"Cheng Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Cheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048730226","display_name":"Jianhao Jiao","orcid":"https://orcid.org/0000-0001-7372-266X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiao, Jianhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127928124","display_name":"Lingyi Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Lingyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068769617","display_name":"Jinqi Xiao","orcid":"https://orcid.org/0009-0004-7311-9413"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao, Jinqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018265542","display_name":"Zhexiang Tang","orcid":"https://orcid.org/0000-0001-6693-0735"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Zhexiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128025647","display_name":"Yu Gong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gong, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120362621","display_name":"Yibiao Ying","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ying, Yibiao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127896017","display_name":"Yang Sui","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sui, Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Lin, Jintian","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Jintian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127943568","display_name":"Wen Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Wen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5127931237","display_name":"Bo Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan, Bo","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":false,"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":0.9574000239372253,"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":0.9574000239372253,"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":0.010700000450015068,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.003700000001117587,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7329000234603882},{"id":"https://openalex.org/keywords/visual-reasoning","display_name":"Visual reasoning","score":0.5996000170707703},{"id":"https://openalex.org/keywords/bounding-overwatch","display_name":"Bounding overwatch","score":0.5322999954223633},{"id":"https://openalex.org/keywords/opportunistic-reasoning","display_name":"Opportunistic reasoning","score":0.49459999799728394},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.4523000121116638},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.4408000111579895},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.4212999939918518},{"id":"https://openalex.org/keywords/abductive-reasoning","display_name":"Abductive reasoning","score":0.41519999504089355}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7329000234603882},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7221999764442444},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6657000184059143},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.5996000170707703},{"id":"https://openalex.org/C63584917","wikidata":"https://www.wikidata.org/wiki/Q333286","display_name":"Bounding overwatch","level":2,"score":0.5322999954223633},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5049999952316284},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.49459999799728394},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.4523000121116638},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.4408000111579895},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.4212999939918518},{"id":"https://openalex.org/C166088908","wikidata":"https://www.wikidata.org/wiki/Q308495","display_name":"Abductive reasoning","level":2,"score":0.41519999504089355},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.4034000039100647},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.39010000228881836},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.3752000033855438},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.3650999963283539},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.31779998540878296},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.30559998750686646},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.30230000615119934},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.30070000886917114},{"id":"https://openalex.org/C193611912","wikidata":"https://www.wikidata.org/wiki/Q4677596","display_name":"Active vision","level":2,"score":0.28279998898506165},{"id":"https://openalex.org/C129916263","wikidata":"https://www.wikidata.org/wiki/Q1141183","display_name":"Backward chaining","level":4,"score":0.2782000005245209},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.26350000500679016},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C3746660","wikidata":"https://www.wikidata.org/wiki/Q1068763","display_name":"Rule of inference","level":2,"score":0.2554999887943268}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.01490","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.01490","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":"doi:10.48550/arxiv.2603.01490","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.01490","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":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Vision-Language-Action":[0],"(VLA)":[1],"models":[2],"rely":[3],"on":[4,55],"current":[5],"observations,":[6],"including":[7],"images,":[8],"language":[9],"instructions,":[10],"and":[11,17,30,70,96,105,129,186,190],"robot":[12],"states,":[13],"to":[14,36,64,84],"predict":[15],"actions":[16],"complete":[18],"tasks.":[19],"While":[20],"accurate":[21],"visual":[22,76,156],"perception":[23],"is":[24,165],"crucial":[25],"for":[26,171],"precise":[27],"action":[28],"prediction":[29],"execution,":[31],"recent":[32],"work":[33],"has":[34],"attempted":[35],"further":[37],"improve":[38],"performance":[39],"by":[40,142],"introducing":[41],"explicit":[42,135],"reasoning":[43,122,140,169],"during":[44],"inference.":[45],"However,":[46],"such":[47,58],"approaches":[48],"face":[49],"significant":[50],"limitations.":[51],"They":[52],"often":[53],"depend":[54],"data-intensive":[56],"resources":[57],"as":[59],"Chain-of-Thought":[60],"(CoT)":[61],"style":[62],"annotations":[63,78],"decompose":[65],"tasks":[66],"into":[67,123],"step-by-step":[68],"reasoning,":[69],"in":[71,101],"many":[72],"cases":[73],"require":[74],"additional":[75],"grounding":[77],"(e.g.,":[79],"bounding":[80],"boxes":[81],"or":[82,134,162],"masks)":[83],"highlight":[85],"relevant":[86],"image":[87],"regions.":[88],"Moreover,":[89],"they":[90],"involve":[91],"time-consuming":[92],"dataset":[93],"construction,":[94],"labeling,":[95],"retraining,":[97],"which":[98],"ultimately":[99],"results":[100],"longer":[102],"inference":[103,125,193],"sequences":[104],"reduced":[106],"efficiency.":[107,194],"To":[108],"address":[109],"these":[110],"challenges,":[111],"we":[112],"propose":[113],"ATA,":[114],"a":[115,166],"novel":[116],"training-free":[117],"framework":[118],"that":[119,180],"introduces":[120],"implicit":[121,168],"VLA":[124,172],"through":[126],"complementary":[127],"attention-guided":[128],"action-guided":[130],"strategies.":[131],"Unlike":[132],"CoT":[133],"visual-grounding":[136],"methods,":[137],"ATA":[138,164],"formulates":[139],"implicitly":[141],"integrating":[143],"attention":[144],"maps":[145],"with":[146],"an":[147],"action-based":[148],"region":[149],"of":[150],"interest":[151],"(RoI),":[152],"thereby":[153],"adaptively":[154],"refining":[155],"inputs":[157],"without":[158],"requiring":[159],"extra":[160],"training":[161],"annotations.":[163],"plug-and-play":[167],"approach":[170],"models,":[173],"lightweight":[174],"yet":[175],"effective.":[176],"Extensive":[177],"experiments":[178],"show":[179],"it":[181],"consistently":[182],"improves":[183],"task":[184],"success":[185],"robustness":[187],"while":[188],"preserving,":[189],"even":[191],"enhancing,":[192]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-04T00:00:00"}