{"id":"https://openalex.org/W7204485930","doi":"https://doi.org/10.48550/arxiv.2608.25350","title":"Beyond Pairwise Feedback: Listwise Vision-Language Supervision for Preference-Based Reward Learning","display_name":"Beyond Pairwise Feedback: Listwise Vision-Language Supervision for Preference-Based Reward Learning","publication_year":2026,"publication_date":"2026-08-26","ids":{"openalex":"https://openalex.org/W7204485930","doi":"https://doi.org/10.48550/arxiv.2608.25350"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.25350","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.25350","pdf_url":"https://arxiv.org/pdf/2608.25350","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.25350","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5148458789","display_name":"Srivalli Katkuri","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Katkuri, Srivalli","raw_affiliation_strings":["School of Mechanical Engineering Purdue University West Lafayette , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Mechanical Engineering Purdue University West Lafayette , USA","institution_ids":["https://openalex.org/I219193219"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148467994","display_name":"Maxwell Kawada","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kawada, Maxwell","raw_affiliation_strings":["Weldon School of Biomedical Engineering Purdue University West Lafayette , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Weldon School of Biomedical Engineering Purdue University West Lafayette , USA","institution_ids":["https://openalex.org/I219193219"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5072528523","display_name":"Juan Pablo Wachs","orcid":"https://orcid.org/0000-0002-6425-5745","observed_orcids":["https://orcid.org/0000-0002-6425-5745"]},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Wachs, Juan","raw_affiliation_strings":["Edwardson School of Industrial Engineering Purdue University West Lafayette , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Edwardson School of Industrial Engineering Purdue University West Lafayette , USA","institution_ids":["https://openalex.org/I219193219"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I219193219"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9984999895095825,"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"}},"topics":[{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.9984999895095825,"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":0.0007999999797903001,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.0,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/vision-language-models","display_name":"vision-language models","score":0.8989999890327454},{"id":"https://openalex.org/keywords/plackett-luce-model","display_name":"Plackett-Luce model","score":0.9369999766349792},{"id":"https://openalex.org/keywords/listwise-ranking","display_name":"listwise ranking","score":0.7260000109672546},{"id":"https://openalex.org/keywords/bradley-terry-model","display_name":"Bradley-Terry model","score":0.8960000276565552},{"id":"https://openalex.org/keywords/reward-modeling","display_name":"reward modeling","score":0.5149999856948853},{"id":"https://openalex.org/keywords/reinforcement-learning-from-human-feedback","display_name":"reinforcement learning from human feedback","score":0.7860000133514404},{"id":"https://openalex.org/keywords/robotic-manipulation","display_name":"robotic manipulation","score":0.6240000128746033},{"id":"https://openalex.org/keywords/meta-world","display_name":"Meta-World","score":0.656000018119812}],"concepts":[{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.9128000140190125},{"id":"https://openalex.org/C181204326","wikidata":"https://www.wikidata.org/wiki/Q7239820","display_name":"Preference learning","level":3,"score":0.7307999730110168},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6972000002861023},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.6920999884605408},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6672000288963318},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.6565999984741211},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6421999931335449},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.6417999863624573},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.636900007724762},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.5038999915122986},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.366100013256073},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.33640000224113464},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.33079999685287476},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.29350000619888306},{"id":"https://openalex.org/C86037889","wikidata":"https://www.wikidata.org/wiki/Q4330127","display_name":"Learning to rank","level":3,"score":0.2883000075817108},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.26190000772476196},{"id":"https://openalex.org/C16345878","wikidata":"https://www.wikidata.org/wiki/Q107472979","display_name":"Orientation (vector space)","level":2,"score":0.25099998712539673}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.25350","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.25350","pdf_url":"https://arxiv.org/pdf/2608.25350","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.25350","doi":"https://doi.org/10.48550/arxiv.2608.25350","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.25350","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.25350","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.25350","pdf_url":"https://arxiv.org/pdf/2608.25350","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G4391784328","display_name":"EAGER: Theoretical Foundations for Integrating Foundational Models into Reinforcement Learning","funder_award_id":"2521982","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320332169","display_name":"Directorate for Computer and Information Science and Engineering","ror":"https://ror.org/025kzpk63"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204485930.pdf","grobid_xml":"https://content.openalex.org/works/W7204485930.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W1588819017","https://openalex.org/W2462906003","https://openalex.org/W3037207827","https://openalex.org/W4226251505","https://openalex.org/W4389821395","https://openalex.org/W4413858428","https://openalex.org/W4416186432","https://openalex.org/W4416780398","https://openalex.org/W4417099698","https://openalex.org/W6948050170","https://openalex.org/W7133242608","https://openalex.org/W7221004523"],"related_works":["https://openalex.org/W4387839182","https://openalex.org/W7166187027","https://openalex.org/W6948050170","https://openalex.org/W4415539998","https://openalex.org/W4407184418","https://openalex.org/W4391631669","https://openalex.org/W4416261729","https://openalex.org/W4416035565","https://openalex.org/W4389821395","https://openalex.org/W7167029937"],"abstract_inverted_index":{"Vision-language":[0],"models":[1,149],"(VLMs)":[2],"have":[3],"emerged":[4],"as":[5,97,157,159],"a":[6,75,91,104,109,244],"powerful":[7],"source":[8,47],"of":[9,27,81,108],"supervision":[10,242],"for":[11,48,103,132,252],"reinforcement":[12,33,253],"learning,":[13],"enabling":[14],"agents":[15],"to":[16,99,116,196,208,249],"leverage":[17],"rich":[18],"semantic":[19],"knowledge":[20],"during":[21],"training.":[22],"Inspired":[23],"by":[24,56],"the":[25,62,121,129,209,228],"success":[26,188,224],"preference-based":[28],"reward":[29,50,92,133,148,250],"learning":[30,34,49,251],"(PbRL)":[31],"in":[32,186],"from":[35,154],"human":[36],"feedback":[37,213],"(RLHF),":[38],"vision-language":[39],"model":[40,93,131],"generated":[41],"image-based":[42],"preferences":[43,127],"provide":[44],"an":[45,220],"effective":[46],"functions.":[51],"This":[52],"can":[53,89,150,204],"be":[54,206],"done":[55],"visually":[57],"comparing":[58],"two":[59,72],"outcomes":[60],"through":[61],"Bradley-Terry":[63],"(BT)":[64],"model.":[65],"However,":[66],"this":[67,114],"pairwise":[68,100,160,191],"formulation":[69,88],"utilizes":[70],"only":[71],"observations":[73],"at":[74,170],"time,":[76],"despite":[77],"VLMs":[78],"being":[79],"capable":[80],"ranking":[82,174,201],"multiple":[83],"candidates.":[84],"The":[85],"Plackett-Luce":[86,130,146],"(PL)":[87,147],"shape":[90],"with":[94,128,181],"listwise":[95,239],"rankings":[96,156],"opposed":[98],"preferences,":[101],"allowing":[102],"more":[105],"suited":[106],"use":[107],"VLM":[110,240],"based":[111],"ranking.":[112],"In":[113],"work,":[115],"our":[117,137],"knowledge,":[118],"we":[119],"introduce":[120],"first":[122],"framework":[123],"that":[124,145,238],"combines":[125],"VLM-generated":[126,155],"learning.":[134,254],"We":[135],"evaluate":[136],"approach":[138,248],"on":[139,231],"Meta-World":[140],"manipulation":[141],"tasks":[142],"and":[143,164,203,211,226,246],"show":[144],"train":[151],"robotic":[152],"policies":[153],"effectively":[158],"Bradley-Terry,":[161,163],"$K$-wise":[162],"RL-VLM-F":[165],"baselines.":[166],"Across":[167],"all":[168],"environments,":[169],"least":[171],"one":[172],"PL":[173,198,217],"size":[175],"($K":[176],"\\in":[177],"\\{3,4,5\\}$)":[178],"consistently":[179],"performs":[180],"or":[182],"outperforms":[183],"other":[184],"methods":[185],"mean":[187,222],"rate.":[189],"Unlike":[190],"methods,":[192],"which":[193],"are":[194],"restricted":[195],"$K=2$,":[197],"supports":[199],"different":[200],"sizes":[202],"therefore":[205],"adapted":[207],"environment":[210],"desired":[212],"format.":[214],"Our":[215],"best":[216],"configuration":[218],"achieves":[219],"86%":[221],"final":[223],"rate":[225],"matches":[227],"Oracle":[229],"baseline":[230],"Drawer":[232],"Open.":[233],"Overall,":[234],"these":[235],"results":[236],"demonstrate":[237],"preference":[241],"is":[243],"competitive":[245],"flexible":[247]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-08-28T00:00:00"}