{"id":"https://openalex.org/W7160432616","doi":"https://doi.org/10.48550/arxiv.2605.03821","title":"RoboAlign-R1: Distilled Multimodal Reward Alignment for Robot Video World Models","display_name":"RoboAlign-R1: Distilled Multimodal Reward Alignment for Robot Video World Models","publication_year":2026,"publication_date":"2026-05-05","ids":{"openalex":"https://openalex.org/W7160432616","doi":"https://doi.org/10.48550/arxiv.2605.03821"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.03821","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.03821","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.2605.03821","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135446185","display_name":"Hao Wu","orcid":"https://orcid.org/0009-0004-9853-8796"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Hao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129701378","display_name":"Yuqi Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yuqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135422336","display_name":"Yuan Gao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135473488","display_name":"Fan Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Fan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135474429","display_name":"Fan Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Fan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135524373","display_name":"Kun Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Kun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135489951","display_name":"Penghao Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Penghao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135484407","display_name":"Qiufeng Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Qiufeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135437241","display_name":"Yizhou Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Yizhou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135430274","display_name":"Weiyan Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Weiyan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074244244","display_name":"Yingli Tian","orcid":"https://orcid.org/0000-0003-4458-360X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tian, Yingli","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135429883","display_name":"Xian Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Xian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5052865881","display_name":"Xiaomeng Huang","orcid":"https://orcid.org/0000-0002-4158-1089"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Xiaomeng","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":0.1746000051498413,"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":0.1746000051498413,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.13109999895095825,"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":0.10400000214576721,"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/robot","display_name":"Robot","score":0.7010999917984009},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.6381000280380249},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5536999702453613},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.54830002784729},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5271999835968018},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.4885999858379364},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.41999998688697815},{"id":"https://openalex.org/keywords/human\u2013robot-interaction","display_name":"Human\u2013robot interaction","score":0.41100001335144043},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.38260000944137573},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.36320000886917114}],"concepts":[{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.7010999917984009},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6930999755859375},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6401000022888184},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.6381000280380249},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5536999702453613},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.54830002784729},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5271999835968018},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.4885999858379364},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.475600004196167},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.42989999055862427},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.41999998688697815},{"id":"https://openalex.org/C145460709","wikidata":"https://www.wikidata.org/wiki/Q859951","display_name":"Human\u2013robot interaction","level":3,"score":0.41100001335144043},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.38260000944137573},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.36320000886917114},{"id":"https://openalex.org/C103910844","wikidata":"https://www.wikidata.org/wiki/Q2631256","display_name":"Video quality","level":3,"score":0.34389999508857727},{"id":"https://openalex.org/C60692881","wikidata":"https://www.wikidata.org/wiki/Q584529","display_name":"Humanoid robot","level":3,"score":0.34380000829696655},{"id":"https://openalex.org/C102392041","wikidata":"https://www.wikidata.org/wiki/Q592860","display_name":"Sliding window protocol","level":3,"score":0.3407000005245209},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.33169999718666077},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3199999928474426},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.3133000135421753},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.29120001196861267},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.28870001435279846},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.27160000801086426},{"id":"https://openalex.org/C104114177","wikidata":"https://www.wikidata.org/wiki/Q79782","display_name":"Motion (physics)","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C19966478","wikidata":"https://www.wikidata.org/wiki/Q4810574","display_name":"Mobile robot","level":3,"score":0.26570001244544983},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.26330000162124634},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.26190000772476196},{"id":"https://openalex.org/C150415221","wikidata":"https://www.wikidata.org/wiki/Q40687","display_name":"Robotic arm","level":2,"score":0.2599000036716461},{"id":"https://openalex.org/C53533937","wikidata":"https://www.wikidata.org/wiki/Q185020","display_name":"Histogram","level":3,"score":0.2581999897956848},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.2574000060558319},{"id":"https://openalex.org/C2779346075","wikidata":"https://www.wikidata.org/wiki/Q7268763","display_name":"Quality Score","level":3,"score":0.2540000081062317},{"id":"https://openalex.org/C207347870","wikidata":"https://www.wikidata.org/wiki/Q371174","display_name":"Gesture","level":2,"score":0.2533999979496002},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2526000142097473}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.03821","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.03821","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.2605.03821","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.03821","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":[{"score":0.6999406218528748,"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Existing":[0],"robot":[1,28,63,80,225],"video":[2,64,226],"world":[3,65,227],"models":[4],"are":[5,18,167],"typically":[6],"trained":[7],"with":[8,21,58,186],"low-level":[9],"objectives":[10],"such":[11],"as":[12],"reconstruction":[13],"and":[14,36,83,159,175,198,211,220],"perceptual":[15],"similarity,":[16],"which":[17],"poorly":[19],"aligned":[20],"the":[22,101,132,142,149],"capabilities":[23],"that":[24,54,129,208],"matter":[25],"most":[26],"for":[27,62,109],"decision":[29],"making,":[30],"including":[31,152],"instruction":[32],"following,":[33],"manipulation":[34],"success,":[35],"physical":[37,218],"plausibility.":[38],"They":[39],"also":[40],"suffer":[41],"from":[42,78],"error":[43],"accumulation":[44],"in":[45,196,202,224],"long-horizon":[46,60,115,183,213,221],"autoregressive":[47],"prediction.":[48],"We":[49,67,98],"present":[50],"RoboAlign-R1,":[51],"a":[52,70,85,104,125,176,193,199],"framework":[53],"combines":[55],"reward-aligned":[56,209],"post-training":[57,210],"stabilized":[59,212],"inference":[61,127],"models.":[66,228],"construct":[68],"RobotWorldBench,":[69],"benchmark":[71],"of":[72,95,154],"10,000":[73],"annotated":[74],"video-instruction":[75],"pairs":[76],"collected":[77],"four":[79],"data":[81],"sources,":[82],"train":[84],"multimodal":[86],"teacher":[87,102],"judge,":[88],"RoboAlign-Judge,":[89],"to":[90],"provide":[91],"fine-grained":[92],"six-dimensional":[93],"evaluation":[94,138],"generated":[96],"videos.":[97],"then":[99],"distill":[100],"into":[103],"lightweight":[105],"student":[106],"reward":[107],"model":[108],"efficient":[110],"reinforcement-learning-based":[111],"post-training.":[112],"To":[113],"reduce":[114],"rollout":[116],"drift,":[117],"we":[118],"further":[119,168],"introduce":[120],"Sliding":[121],"Window":[122],"Re-encoding":[123],"(SWR),":[124],"training-free":[126],"strategy":[128],"periodically":[130],"refreshes":[131],"generation":[133],"context.":[134],"Under":[135],"our":[136],"in-domain":[137],"protocol,":[139],"RoboAlign-R1":[140],"improves":[141,182],"aggregate":[143],"six-dimension":[144],"score":[145],"by":[146,170],"10.1%":[147],"over":[148],"strongest":[150],"baseline,":[151],"gains":[153],"7.5%":[155],"on":[156,161],"Manipulation":[157],"Accuracy":[158],"4.6%":[160],"Instruction":[162],"Following;":[163],"these":[164,205],"ranking":[165],"improvements":[166],"supported":[169],"an":[171],"external":[172],"VLM-based":[173],"cross-check":[174],"blinded":[177],"human":[178],"study.":[179],"Meanwhile,":[180],"SWR":[181],"prediction":[184,222],"quality":[185,223],"only":[187],"about":[188],"1%":[189],"additional":[190],"latency,":[191],"yielding":[192],"2.8%":[194],"gain":[195],"SSIM":[197],"9.8%":[200],"reduction":[201],"LPIPS.":[203],"Together,":[204],"results":[206],"show":[207],"decoding":[214],"improve":[215],"task":[216],"consistency,":[217],"realism,":[219]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-07T00:00:00"}