{"id":"https://openalex.org/W7163711616","doi":"https://doi.org/10.48550/arxiv.2606.06245","title":"MPCoT: Reward-Guided Multi-Path Latent Reasoning for Test-Time Scalable Vision-Language-Action","display_name":"MPCoT: Reward-Guided Multi-Path Latent Reasoning for Test-Time Scalable Vision-Language-Action","publication_year":2026,"publication_date":"2026-06-04","ids":{"openalex":"https://openalex.org/W7163711616","doi":"https://doi.org/10.48550/arxiv.2606.06245"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.06245","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06245","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":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.2606.06245","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137991660","display_name":"Boyang Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Boyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5010380116","display_name":"Lianlei Shan","orcid":"https://orcid.org/0000-0002-4648-8246"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shan, Lianlei","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":0.9282000064849854,"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":0.9282000064849854,"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":0.01489999983459711,"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":0.008100000210106373,"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/path","display_name":"Path (computing)","score":0.6579999923706055},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6140000224113464},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.5910000205039978},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5527999997138977},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.5227000117301941},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5203999876976013}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.757099986076355},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.6579999923706055},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6140000224113464},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.5910000205039978},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5527999997138977},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5227000117301941},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5203999876976013},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.4902999997138977},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44339999556541443},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2989000082015991},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.2888000011444092},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.28790000081062317},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2750000059604645},{"id":"https://openalex.org/C124978682","wikidata":"https://www.wikidata.org/wiki/Q1201019","display_name":"Proof of concept","level":2,"score":0.26460000872612},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.2587999999523163}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.06245","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06245","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":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.2606.06245","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06245","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.48367196321487427,"id":"https://metadata.un.org/sdg/16"}],"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],"policies":[2],"remain":[3],"brittle":[4],"in":[5],"long-horizon":[6,111],"and":[7,28,52,72,96,107,120],"high-uncertainty":[8],"control,":[9],"where":[10],"one-pass":[11],"action":[12,57,65,90],"decoding":[13],"provides":[14],"limited":[15],"inference-time":[16],"deliberation.":[17],"Explicit":[18],"chain-of-thought":[19],"can":[20],"increase":[21],"reasoning":[22,40,94],"depth,":[23],"but":[24],"introduces":[25],"token":[26],"latency":[27],"an":[29],"indirect":[30],"text-to-action":[31],"interface.":[32],"We":[33],"propose":[34],"MPCoT,":[35],"a":[36],"reward-guided":[37,121],"multi-path":[38],"latent":[39,78],"framework":[41],"that":[42],"initializes":[43],"$M$":[44],"hypotheses,":[45],"refines":[46],"them":[47,55],"for":[48],"K":[49],"weight-tied":[50],"steps,":[51],"softly":[53],"aggregates":[54],"before":[56],"decoding.":[58],"A":[59],"training-only":[60],"path-preference":[61],"objective":[62],"evaluates":[63],"candidate":[64],"branches":[66],"with":[67,81,113],"expert-action":[68],"consistency,":[69],"world-model/VLM-based":[70],"progress,":[71],"success":[73],"feedback":[74],"to":[75],"align":[76],"the":[77,87],"path":[79,122],"scorer":[80],"downstream":[82],"execution":[83],"quality.":[84],"MPCoT":[85,109],"preserves":[86],"original":[88],"8-step":[89],"interface,":[91],"generates":[92],"zero":[93],"tokens,":[95],"exposes":[97],"configurable":[98],"inference":[99],"controls":[100],"(K,M).":[101],"Under":[102],"matched":[103],"protocols":[104],"on":[105],"LIBERO":[106],"CALVIN,":[108],"improves":[110],"performance,":[112],"ablations":[114],"confirming":[115],"depth-width":[116],"effects,":[117],"confidence-weighted":[118],"aggregation,":[119],"supervision.":[123]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-06T00:00:00"}