{"id":"https://openalex.org/W7128539365","doi":"https://doi.org/10.48550/arxiv.2602.07541","title":"Differentiate-and-Inject: Enhancing VLAs via Functional Differentiation Induced by In-Parameter Structural Reasoning","display_name":"Differentiate-and-Inject: Enhancing VLAs via Functional Differentiation Induced by In-Parameter Structural Reasoning","publication_year":2026,"publication_date":"2026-02-07","ids":{"openalex":"https://openalex.org/W7128539365","doi":"https://doi.org/10.48550/arxiv.2602.07541"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.07541","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"article","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5012800942","display_name":"Jingyi Hou","orcid":"https://orcid.org/0000-0003-0172-3115"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hou, Jingyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125494077","display_name":"Leyu Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Leyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125490154","display_name":"Chenchen Jing","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jing, Chenchen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125489996","display_name":"Jinghan Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Jinghan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021537732","display_name":"Xinbo Yu","orcid":"https://orcid.org/0000-0003-0958-6020"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Xinbo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5125528424","display_name":"Wei He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Wei","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":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.17470478,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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":0.8428999781608582,"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":0.8428999781608582,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.05979999899864197,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.021800000220537186,"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/task","display_name":"Task (project management)","score":0.777899980545044},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.7480999827384949},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7225000262260437},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.5105000138282776},{"id":"https://openalex.org/keywords/automated-reasoning","display_name":"Automated reasoning","score":0.3833000063896179},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.3151000142097473}],"concepts":[{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.777899980545044},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7480999827384949},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7225000262260437},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6948999762535095},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.603600025177002},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.5105000138282776},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.3833000063896179},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3151000142097473},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3095000088214874},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2890999913215637},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.2680000066757202},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.26759999990463257},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2667999863624573},{"id":"https://openalex.org/C47701112","wikidata":"https://www.wikidata.org/wiki/Q735188","display_name":"Protein structure","level":2,"score":0.26010000705718994}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.07541","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.07541","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.07541","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":"pmh:doi:10.48550/arxiv.2602.07541","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"As":[0],"robots":[1],"are":[2],"expected":[3],"to":[4,51],"perform":[5],"increasingly":[6],"diverse":[7,141],"tasks,":[8],"they":[9],"must":[10],"understand":[11],"not":[12],"only":[13],"low-level":[14,67],"actions":[15],"but":[16],"also":[17],"the":[18,121,162],"higher-level":[19],"structure":[20,101,119],"that":[21,128],"determines":[22],"how":[23],"a":[24,76],"task":[25,73,148,175],"should":[26],"unfold.":[27],"Existing":[28],"vision-language-action":[29],"(VLA)":[30],"models":[31,81],"struggle":[32],"with":[33,66],"this":[34],"form":[35,122],"of":[36,91,123,164],"task-level":[37,64,99,108,135],"reasoning.":[38,89],"They":[39],"either":[40],"depend":[41],"on":[42],"prompt-based":[43],"in-context":[44,156],"decomposition,":[45],"which":[46,58],"is":[47],"unstable":[48],"and":[49,62,134,150,157,171],"sensitive":[50],"linguistic":[52],"variations,":[53],"or":[54,113],"end-to-end":[55,158],"long-horizon":[56],"training,":[57],"requires":[59],"large-scale":[60],"demonstrations":[61],"entangles":[63],"reasoning":[65,74,167],"control.":[68],"We":[69],"present":[70],"in-parameter":[71,87],"structured":[72],"(iSTAR),":[75],"framework":[77],"for":[78,168],"enhancing":[79],"VLA":[80,159],"via":[82],"functional":[83,169],"differentiation":[84,170],"induced":[85],"by":[86],"structural":[88,166],"Instead":[90],"treating":[92],"VLAs":[93],"as":[94],"monolithic":[95],"policies,":[96],"iSTAR":[97,144],"embeds":[98],"semantic":[100],"directly":[102],"into":[103],"model":[104],"parameters,":[105],"enabling":[106],"differentiated":[107],"inference":[109],"without":[110],"external":[111],"planners":[112],"handcrafted":[114],"prompt":[115],"inputs.":[116],"This":[117],"injected":[118],"takes":[120],"implicit":[124],"dynamic":[125],"scene-graph":[126],"knowledge":[127],"captures":[129],"object":[130],"relations,":[131],"subtask":[132],"semantics,":[133],"dependencies":[136],"in":[137],"parameter":[138],"space.":[139],"Across":[140],"manipulation":[142],"benchmarks,":[143],"achieves":[145],"more":[146],"reliable":[147],"decompositions":[149],"higher":[151],"success":[152],"rates":[153],"than":[154],"both":[155],"baselines,":[160],"demonstrating":[161],"effectiveness":[163],"parameter-space":[165],"improved":[172],"generalization":[173],"across":[174],"variations.":[176]},"counts_by_year":[],"updated_date":"2026-07-15T18:14:33.161393","created_date":"2026-02-11T00:00:00"}