{"id":"https://openalex.org/W4416748544","doi":"https://doi.org/10.1109/iros60139.2025.11246902","title":"Policy Learning from Large Vision-Language Model Feedback Without Reward Modeling","display_name":"Policy Learning from Large Vision-Language Model Feedback Without Reward Modeling","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4416748544","doi":"https://doi.org/10.1109/iros60139.2025.11246902"},"language":null,"primary_location":{"id":"doi:10.1109/iros60139.2025.11246902","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros60139.2025.11246902","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5028393159","display_name":"Tung M. Luu","orcid":"https://orcid.org/0000-0001-9488-7463"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Tung M. Luu","raw_affiliation_strings":["KAIST (Korea Advanced Institute of Science and Technology),School of Electrical Engineering,Daejeon,Republic of Korea"],"affiliations":[{"raw_affiliation_string":"KAIST (Korea Advanced Institute of Science and Technology),School of Electrical Engineering,Daejeon,Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101787061","display_name":"Dong-Hoon Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Donghoon Lee","raw_affiliation_strings":["KAIST (Korea Advanced Institute of Science and Technology),School of Electrical Engineering,Daejeon,Republic of Korea"],"affiliations":[{"raw_affiliation_string":"KAIST (Korea Advanced Institute of Science and Technology),School of Electrical Engineering,Daejeon,Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101422405","display_name":"Young-Hwan Lee","orcid":"https://orcid.org/0009-0002-2310-3056"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Younghwan Lee","raw_affiliation_strings":["KAIST (Korea Advanced Institute of Science and Technology),School of Electrical Engineering,Daejeon,Republic of Korea"],"affiliations":[{"raw_affiliation_string":"KAIST (Korea Advanced Institute of Science and Technology),School of Electrical Engineering,Daejeon,Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5073287748","display_name":"Chang D. Yoo","orcid":"https://orcid.org/0000-0002-0756-7179"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Chang D. Yoo","raw_affiliation_strings":["KAIST (Korea Advanced Institute of Science and Technology),School of Electrical Engineering,Daejeon,Republic of Korea"],"affiliations":[{"raw_affiliation_string":"KAIST (Korea Advanced Institute of Science and Technology),School of Electrical Engineering,Daejeon,Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5028393159"],"corresponding_institution_ids":["https://openalex.org/I157485424"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.40018631,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11685","last_page":"11692"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.7732999920845032,"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":0.7732999920845032,"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":0.08609999716281891,"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":0.05469999834895134,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.7067999839782715},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6710000038146973},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.5546000003814697},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.5332000255584717},{"id":"https://openalex.org/keywords/preference-learning","display_name":"Preference learning","score":0.5092999935150146},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.4643000066280365},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.44190001487731934},{"id":"https://openalex.org/keywords/task-analysis","display_name":"Task analysis","score":0.37459999322891235},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.3353999853134155}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7756999731063843},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7067999839782715},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6710000038146973},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6381999850273132},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.5546000003814697},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5501999855041504},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.5332000255584717},{"id":"https://openalex.org/C181204326","wikidata":"https://www.wikidata.org/wiki/Q7239820","display_name":"Preference learning","level":3,"score":0.5092999935150146},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.4643000066280365},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.44190001487731934},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.37459999322891235},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3353999853134155},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.31940001249313354},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.31619998812675476},{"id":"https://openalex.org/C133462117","wikidata":"https://www.wikidata.org/wiki/Q4929239","display_name":"Data collection","level":2,"score":0.3077000081539154},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.3000999987125397},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.29510000348091125},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.2948000133037567},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.289900004863739},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.26969999074935913},{"id":"https://openalex.org/C2986087404","wikidata":"https://www.wikidata.org/wiki/Q15946010","display_name":"Online learning","level":2,"score":0.2581000030040741},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.25360000133514404},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.2515999972820282},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.25099998712539673}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iros60139.2025.11246902","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros60139.2025.11246902","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":18,"referenced_works":["https://openalex.org/W2964273174","https://openalex.org/W2982316857","https://openalex.org/W2997423597","https://openalex.org/W3107153805","https://openalex.org/W3141423234","https://openalex.org/W4283705832","https://openalex.org/W4310705874","https://openalex.org/W4383097638","https://openalex.org/W4383108457","https://openalex.org/W4385430679","https://openalex.org/W4385572799","https://openalex.org/W4386285856","https://openalex.org/W4389523762","https://openalex.org/W4390872513","https://openalex.org/W4391759936","https://openalex.org/W4403021114","https://openalex.org/W4405785192","https://openalex.org/W4412158322"],"related_works":[],"abstract_inverted_index":{"Offline":[0],"reinforcement":[1],"learning":[2,134],"(RL)":[3],"provides":[4],"a":[5,76,102,115,130,180],"powerful":[6],"framework":[7],"for":[8,19,89,104],"training":[9],"robotic":[10,148],"agents":[11],"using":[12,129],"pre-collected,":[13],"suboptimal":[14],"datasets,":[15],"eliminating":[16],"the":[17,137,152,171],"need":[18,138],"costly,":[20,63],"time-consuming,":[21],"and":[22,41,65],"potentially":[23],"hazardous":[24],"online":[25,36],"interactions.":[26],"This":[27],"is":[28,39,61,121],"particularly":[29],"useful":[30],"in":[31,175],"safety-critical":[32],"real-world":[33,176],"applications,":[34],"where":[35],"data":[37],"collection":[38],"expensive":[40],"impractical.":[42],"However,":[43],"existing":[44,162],"offline":[45],"RL":[46],"algorithms":[47],"typically":[48],"require":[49],"reward":[50,58,98,142,165],"labeled":[51],"data,":[52],"which":[53],"introduces":[54],"an":[55],"additional":[56],"bottleneck:":[57],"function":[59],"design":[60],"itself":[62],"labor-intensive,":[64],"requires":[66],"significant":[67],"domain":[68],"expertise.":[69],"In":[70],"this":[71],"paper,":[72],"we":[73,169],"introduce":[74],"PLARE,":[75],"novel":[77],"approach":[78],"that":[79],"leverages":[80],"large":[81],"vision-language":[82],"models":[83],"(VLMs)":[84],"to":[85,139],"provide":[86],"guidance":[87],"signals":[88],"agent":[90],"training.":[91],"Instead":[92],"of":[93,109,173],"relying":[94],"on":[95,107,114,147,157],"manually":[96],"designed":[97],"functions,":[99],"PLARE":[100,154,174],"queries":[101],"VLM":[103],"preference":[105,127,133],"labels":[106,128],"pairs":[108],"visual":[110],"trajectory":[111],"segments":[112],"based":[113],"language":[116],"task":[117],"description.":[118],"The":[119],"policy":[120],"then":[122],"trained":[123],"directly":[124],"from":[125,151],"these":[126],"supervised":[131],"contrastive":[132],"objective,":[135],"bypassing":[136],"learn":[140],"explicit":[141],"models.":[143],"Through":[144],"extensive":[145],"experiments":[146],"manipulation":[149,177],"tasks":[150,178],"MetaWorld,":[153],"achieves":[155],"performance":[156],"par":[158],"with":[159,179],"or":[160],"surpassing":[161],"state-of-the-art":[163],"VLM-based":[164],"generation":[166],"methods.":[167],"Furthermore,":[168],"demonstrate":[170],"effectiveness":[172],"physical":[181],"robot,":[182],"further":[183],"validating":[184],"its":[185],"practical":[186],"applicability.":[187]},"counts_by_year":[],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-11-28T00:00:00"}