{"id":"https://openalex.org/W4408352916","doi":"https://doi.org/10.1109/icassp49660.2025.10889042","title":"Reward Generation via Large Vision-Language Model in Offline Reinforcement Learning","display_name":"Reward Generation via Large Vision-Language Model in Offline Reinforcement Learning","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408352916","doi":"https://doi.org/10.1109/icassp49660.2025.10889042"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10889042","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889042","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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/A5072296235","display_name":"Younghwan 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":true,"raw_author_name":"Younghwan Lee","raw_affiliation_strings":["KAIST,Electrical Engineering,Daejeon,South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST,Electrical Engineering,Daejeon,South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","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":false,"raw_author_name":"Tung M. Luu","raw_affiliation_strings":["KAIST,Electrical Engineering,Daejeon,South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST,Electrical Engineering,Daejeon,South Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100422966","display_name":"Donghoon Lee","orcid":"https://orcid.org/0000-0002-2886-2046"},"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,Robotics Program,Daejeon,South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST,Robotics Program,Daejeon,South 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,Electrical Engineering,Daejeon,South Korea"],"affiliations":[{"raw_affiliation_string":"KAIST,Electrical Engineering,Daejeon,South Korea","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5072296235"],"corresponding_institution_ids":["https://openalex.org/I157485424"],"apc_list":null,"apc_paid":null,"fwci":2.1819,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.87518429,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.4912000000476837,"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":0.4912000000476837,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.8374329805374146},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7885404229164124},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5364331603050232},{"id":"https://openalex.org/keywords/reinforcement","display_name":"Reinforcement","score":0.42425450682640076},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.38322150707244873},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.34455716609954834},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.09885707497596741}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8374329805374146},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7885404229164124},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5364331603050232},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.42425450682640076},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38322150707244873},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.34455716609954834},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.09885707497596741},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10889042","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10889042","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","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":33,"referenced_works":["https://openalex.org/W3025019304","https://openalex.org/W3034758614","https://openalex.org/W3039737909","https://openalex.org/W3127561923","https://openalex.org/W3141423234","https://openalex.org/W3152878473","https://openalex.org/W4285218426","https://openalex.org/W4360584316","https://openalex.org/W4389666460","https://openalex.org/W4392679398","https://openalex.org/W4401024730","https://openalex.org/W4401413695","https://openalex.org/W4402727764","https://openalex.org/W4405785192","https://openalex.org/W6740801417","https://openalex.org/W6743368274","https://openalex.org/W6757469721","https://openalex.org/W6779265984","https://openalex.org/W6786103492","https://openalex.org/W6791353385","https://openalex.org/W6796168157","https://openalex.org/W6809509765","https://openalex.org/W6810334672","https://openalex.org/W6810640255","https://openalex.org/W6811013733","https://openalex.org/W6849861922","https://openalex.org/W6851787687","https://openalex.org/W6857415689","https://openalex.org/W6857462693","https://openalex.org/W6857844089","https://openalex.org/W6858048387","https://openalex.org/W6861420795","https://openalex.org/W6866278122"],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W2772917594","https://openalex.org/W2036807459","https://openalex.org/W2058170566","https://openalex.org/W2755342338","https://openalex.org/W2166024367","https://openalex.org/W3116076068","https://openalex.org/W2229312674","https://openalex.org/W2951359407"],"abstract_inverted_index":{"In":[0],"offline":[1,31,90],"reinforcement":[2],"learning":[3,5,41],"(RL),":[4],"from":[6,89],"fixed":[7],"datasets":[8],"presents":[9],"a":[10],"promising":[11],"solution":[12],"for":[13,30],"domains":[14],"where":[15],"real-time":[16],"interaction":[17],"with":[18,42,106],"the":[19,57,81,107],"environment":[20],"is":[21],"expensive":[22],"or":[23],"risky.":[24],"However,":[25],"designing":[26],"dense":[27],"reward":[28,64,109,121],"signals":[29,110],"dataset":[32],"requires":[33],"significant":[34],"human":[35,43,93],"effort":[36],"and":[37,101],"domain":[38],"expertise.":[39],"Reinforcement":[40],"feedback":[44],"(RLHF)":[45],"has":[46],"emerged":[47],"as":[48,118],"an":[49,119],"alternative,":[50],"but":[51],"it":[52],"remains":[53],"costly":[54],"due":[55],"to":[56,86,111],"human-in-the-loop":[58],"process,":[59],"prompting":[60],"interest":[61],"in":[62,98],"automated":[63],"generation":[65],"models.":[66],"To":[67],"address":[68],"this,":[69],"we":[70],"propose":[71],"Reward":[72],"Generation":[73],"via":[74],"Large":[75],"Vision-Language":[76],"Models":[77],"(RG-VLM),":[78],"which":[79],"leverages":[80],"reasoning":[82],"capabilities":[83],"of":[84],"LVLMs":[85],"generate":[87],"rewards":[88],"data":[91],"without":[92],"involvement.":[94],"RG-VLM":[95],"improves":[96],"generalization":[97],"long-horizon":[99],"tasks":[100],"can":[102],"be":[103],"seamlessly":[104],"integrated":[105],"sparse":[108],"enhance":[112],"task":[113],"performance,":[114],"demonstrating":[115],"its":[116],"potential":[117],"auxiliary":[120],"signal.":[122]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-04-23T09:07:50.710637","created_date":"2025-10-10T00:00:00"}