{"id":"https://openalex.org/W7161237890","doi":"https://doi.org/10.48550/arxiv.2605.14274","title":"CreFlow: Corrective Reflow for Sparse-Reward Embodied Video Diffusion RL","display_name":"CreFlow: Corrective Reflow for Sparse-Reward Embodied Video Diffusion RL","publication_year":2026,"publication_date":"2026-05-14","ids":{"openalex":"https://openalex.org/W7161237890","doi":"https://doi.org/10.48550/arxiv.2605.14274"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.14274","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14274","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.14274","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136273671","display_name":"Zhenyang Ni","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ni, Zhenyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136237623","display_name":"Yijiang Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yijiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017515613","display_name":"Ruochen Jiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiao, Ruochen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026560345","display_name":"Simon Sinong Zhan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhan, Simon Sinong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136240971","display_name":"Sipeng Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Sipeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136214334","display_name":"Zhenfei Yin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yin, Zhenfei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136198783","display_name":"Minshuo Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Minshuo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136245975","display_name":"Philip Torr","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Torr, Philip","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136216084","display_name":"Zhaoran Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zhaoran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136247713","display_name":"Qi Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Qi","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/T10653","display_name":"Robot Manipulation and Learning","score":0.31949999928474426,"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"}},"topics":[{"id":"https://openalex.org/T10653","display_name":"Robot Manipulation and Learning","score":0.31949999928474426,"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"}},{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.13819999992847443,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.11190000176429749,"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/task","display_name":"Task (project management)","score":0.7404000163078308},{"id":"https://openalex.org/keywords/embodied-cognition","display_name":"Embodied cognition","score":0.7206000089645386},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.6025000214576721},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.4147999882698059},{"id":"https://openalex.org/keywords/online-model","display_name":"Online model","score":0.36890000104904175},{"id":"https://openalex.org/keywords/task-analysis","display_name":"Task analysis","score":0.3456999957561493}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7684000134468079},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.7404000163078308},{"id":"https://openalex.org/C100609095","wikidata":"https://www.wikidata.org/wiki/Q1335050","display_name":"Embodied cognition","level":2,"score":0.7206000089645386},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.6025000214576721},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.4147999882698059},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38339999318122864},{"id":"https://openalex.org/C2777851325","wikidata":"https://www.wikidata.org/wiki/Q7094102","display_name":"Online model","level":2,"score":0.36890000104904175},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.357699990272522},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.34779998660087585},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3456999957561493},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.3264999985694885},{"id":"https://openalex.org/C2775960376","wikidata":"https://www.wikidata.org/wiki/Q1435859","display_name":"Grippers","level":2,"score":0.32589998841285706},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.32170000672340393},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.27790001034736633},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2741999924182892},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.2639999985694885}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.14274","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14274","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.14274","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14274","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":"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":{"Video":[0],"generation":[1,75,107],"models":[2],"trained":[3],"on":[4],"heterogeneous":[5],"data":[6],"with":[7,121,177],"likelihood-surrogate":[8],"objectives":[9],"can":[10],"produce":[11],"visually":[12],"plausible":[13],"rollouts":[14],"that":[15,130,151,170],"violate":[16],"physical":[17],"constraints":[18],"in":[19,97,104],"embodied":[20,73],"manipulation.":[21],"Although":[22],"reinforcement-learning":[23],"post-training":[24,72],"offers":[25],"a":[26,40,56,66,83,116,126,147],"natural":[27],"route":[28],"to":[29,39,135,140],"adapting":[30],"VGMs,":[31],"existing":[32,184],"video-RL":[33],"rewards":[34,92],"often":[35],"reduce":[36],"each":[37],"rollout":[38,54],"low-level":[41],"visual":[42],"metric,":[43],"whereas":[44],"manipulation":[45,198],"video":[46,74,106],"evaluation":[47],"requires":[48],"logic-based":[49],"verification":[50],"of":[51,85,160],"whether":[52],"the":[53,132,161],"satisfies":[55],"compositional":[57,67],"task":[58,80],"specification.":[59],"To":[60,100],"fill":[61],"this":[62],"gap,":[63],"we":[64,112],"introduce":[65],"constraint-based":[68],"reward":[69,110,173],"model":[70],"for":[71],"models,":[76],"which":[77],"automatically":[78],"formulates":[79],"requirements":[81],"as":[82,156],"composition":[84],"Linear":[86],"Temporal":[87],"Logic":[88],"constraints,":[89],"providing":[90],"faithful":[91],"and":[93,145,165,179,186],"localized":[94],"error":[95],"information":[96],"generated":[98],"videos.":[99],"achieve":[101],"effective":[102],"improvement":[103],"high-dimensional":[105],"using":[108],"these":[109],"signals,":[111],"further":[113],"propose":[114],"CreFlow,":[115],"novel":[117],"online":[118],"RL":[119,133],"framework":[120],"two":[122],"key":[123],"designs:":[124],"i)":[125],"credit-aware":[127],"NFT":[128],"loss":[129,150],"confines":[131],"update":[134],"reward-relevant":[136],"regions,":[137],"preventing":[138],"perturbations":[139],"unrelated":[141],"regions":[142],"during":[143],"post-training;":[144],"ii)":[146],"corrective":[148],"reflow":[149],"leverages":[152],"within-group":[153],"positive":[154],"samples":[155],"an":[157],"explicit":[158],"estimate":[159],"correction":[162],"direction,":[163],"stabilizing":[164],"accelerating":[166],"training.":[167],"Experiments":[168],"show":[169],"CreFlow":[171],"yields":[172],"judgments":[174],"better":[175],"aligned":[176],"human":[178],"simulator":[180],"success":[181,190],"labels":[182],"than":[183],"methods":[185],"improves":[187],"downstream":[188],"execution":[189],"by":[191],"23.8":[192],"percentage":[193],"points":[194],"across":[195],"eight":[196],"bimanual":[197],"tasks.":[199]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-16T00:00:00"}