{"id":"https://openalex.org/W7140229392","doi":"https://doi.org/10.48550/arxiv.2603.21341","title":"RoboAlign: Learning Test-Time Reasoning for Language-Action Alignment in Vision-Language-Action Models","display_name":"RoboAlign: Learning Test-Time Reasoning for Language-Action Alignment in Vision-Language-Action Models","publication_year":2026,"publication_date":"2026-03-22","ids":{"openalex":"https://openalex.org/W7140229392","doi":"https://doi.org/10.48550/arxiv.2603.21341"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.21341","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.21341","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.21341","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Kim, Dongyoung","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Dongyoung","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Park, Sumin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Park, Sumin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Song, Woomin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Woomin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Kim, Seungku","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Seungku","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Kim, Taeyoung","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Taeyoung","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Jang, Huiwon","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jang, Huiwon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Shin, Jinwoo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shin, Jinwoo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Kim, Jaehyung","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kim, Jaehyung","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Seo, Younggyo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Seo, Younggyo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":9,"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":0.9794999957084656,"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":0.9794999957084656,"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":0.0032999999821186066,"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/T10709","display_name":"Social Robot Interaction and HRI","score":0.00139999995008111,"subfield":{"id":"https://openalex.org/subfields/3207","display_name":"Social Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.592199981212616},{"id":"https://openalex.org/keywords/embodied-cognition","display_name":"Embodied cognition","score":0.48809999227523804},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4821000099182129},{"id":"https://openalex.org/keywords/modality","display_name":"Modality (human\u2013computer interaction)","score":0.4657000005245209},{"id":"https://openalex.org/keywords/robotics","display_name":"Robotics","score":0.45350000262260437},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.43639999628067017},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.43369999527931213},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.4108999967575073},{"id":"https://openalex.org/keywords/commonsense-reasoning","display_name":"Commonsense reasoning","score":0.4041000008583069},{"id":"https://openalex.org/keywords/cognitive-robotics","display_name":"Cognitive robotics","score":0.4027999937534332}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.748199999332428},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7293000221252441},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.592199981212616},{"id":"https://openalex.org/C100609095","wikidata":"https://www.wikidata.org/wiki/Q1335050","display_name":"Embodied cognition","level":2,"score":0.48809999227523804},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4821000099182129},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.4657000005245209},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.45350000262260437},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.43639999628067017},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.43369999527931213},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.43230000138282776},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.4108999967575073},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.4041000008583069},{"id":"https://openalex.org/C192327766","wikidata":"https://www.wikidata.org/wiki/Q1038799","display_name":"Cognitive robotics","level":3,"score":0.4027999937534332},{"id":"https://openalex.org/C2983448237","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Language understanding","level":2,"score":0.39100000262260437},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.37560001015663147},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.3580000102519989},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.35350000858306885},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.34619998931884766},{"id":"https://openalex.org/C183759332","wikidata":"https://www.wikidata.org/wiki/Q343680","display_name":"Action learning","level":4,"score":0.33730000257492065},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.3314000070095062},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.3287000060081482},{"id":"https://openalex.org/C161301231","wikidata":"https://www.wikidata.org/wiki/Q3478658","display_name":"Knowledge representation and reasoning","level":2,"score":0.3206000030040741},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.31520000100135803},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.3138999938964844},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.3100000023841858},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.3034999966621399},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.28619998693466187},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.2824000120162964},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.2761000096797943},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.27230000495910645},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2694999873638153},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.2639999985694885},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2614000141620636}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.21341","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.21341","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.21341","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.21341","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"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":{"Improving":[0],"embodied":[1,32],"reasoning":[2,33,90,94],"in":[3,34,49,116],"multimodal-large-language":[4],"models":[5,12],"(MLLMs)":[6],"is":[7,81],"essential":[8],"for":[9],"building":[10],"vision-language-action":[11],"(VLAs)":[13],"on":[14,141,150,180],"top":[15,142],"of":[16,38,130,143,165,172],"them":[17,149],"to":[18,47,82,99,124],"readily":[19],"translate":[20],"multimodal":[21],"understanding":[22],"into":[23],"low-level":[24,114],"actions.":[25],"Accordingly,":[26],"recent":[27],"work":[28],"has":[29],"explored":[30],"enhancing":[31],"MLLMs":[35],"through":[36],"supervision":[37],"vision-question-answering":[39],"type.":[40],"However,":[41],"these":[42],"approaches":[43],"have":[44],"been":[45],"reported":[46],"result":[48],"unstable":[50],"VLA":[51,76],"performance,":[52],"often":[53],"yielding":[54],"only":[55],"marginal":[56],"or":[57],"even":[58],"negative":[59],"gains.":[60],"In":[61],"this":[62,93],"paper,":[63],"we":[64,132],"propose":[65],"a":[66,104,137],"more":[67],"systematic":[68],"MLLM":[69,123,145],"training":[70],"framework":[71],"RoboAlign":[72,106,168],"that":[73],"reliably":[74],"improves":[75],"performance.":[77],"Our":[78],"key":[79],"idea":[80],"sample":[83],"action":[84,101,139],"tokens":[85],"via":[86],"zero-shot":[87],"natural":[88],"language":[89,112],"and":[91,113,118,147,175,183],"refines":[92],"using":[95,161],"reinforcement":[96],"learning":[97],"(RL)":[98],"improve":[100],"accuracy.":[102],"As":[103],"result,":[105],"bridges":[107],"the":[108,128,166],"modality":[109],"gap":[110],"between":[111],"actions":[115],"MLLMs,":[117],"facilitate":[119],"knowledge":[120],"transfer":[121],"from":[122],"VLA.":[125],"To":[126],"validate":[127],"effectiveness":[129],"RoboAlign,":[131],"train":[133],"VLAs":[134],"by":[135,155],"adding":[136],"diffusion-based":[138],"head":[140],"an":[144],"backbone":[146],"evaluate":[148],"major":[151],"robotics":[152],"benchmarks.":[153],"Remarkably,":[154],"performing":[156],"RL-based":[157],"alignment":[158],"after":[159],"SFT":[160,178],"less":[162],"than":[163],"1\\%":[164],"data,":[167],"achieves":[169],"performance":[170],"improvements":[171],"17.5\\%,":[173],"18.9\\%,":[174],"106.6\\%":[176],"over":[177],"baselines":[179],"LIBERO,":[181],"CALVIN,":[182],"real-world":[184],"environments,":[185],"respectively.":[186]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-25T00:00:00"}
