{"id":"https://openalex.org/W4386437460","doi":"https://doi.org/10.48550/arxiv.2309.00230","title":"JoTR: A Joint Transformer and Reinforcement Learning Framework for Dialog Policy Learning","display_name":"JoTR: A Joint Transformer and Reinforcement Learning Framework for Dialog Policy Learning","publication_year":2023,"publication_date":"2023-09-01","ids":{"openalex":"https://openalex.org/W4386437460","doi":"https://doi.org/10.48550/arxiv.2309.00230"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2309.00230","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2309.00230","pdf_url":"https://arxiv.org/pdf/2309.00230","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2309.00230","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5068169387","display_name":"Wai-Chung Kwan","orcid":"https://orcid.org/0000-0002-2942-4208","observed_orcids":["https://orcid.org/0000-0002-2942-4208"]},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Kwan, Wai-Chung","raw_affiliation_strings":["The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100411037","display_name":"Huimin Wang","orcid":"https://orcid.org/0009-0006-3932-1604","observed_orcids":["https://orcid.org/0009-0006-3932-1604"]},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]},{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN","HK"],"is_corresponding":false,"raw_author_name":"Wang, Huimin","raw_affiliation_strings":["Jarvis Lab , Tencent","The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jarvis Lab , Tencent","institution_ids":["https://openalex.org/I2250653659"]},{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075179865","display_name":"Hongru Wang","orcid":"https://orcid.org/0000-0001-8305-5231","observed_orcids":["https://orcid.org/0000-0001-8305-5231"]},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]},{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN","HK"],"is_corresponding":false,"raw_author_name":"Wang, Hongru","raw_affiliation_strings":["Jarvis Lab , Tencent","The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jarvis Lab , Tencent","institution_ids":["https://openalex.org/I2250653659"]},{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100747388","display_name":"Zezhong Wang","orcid":"https://orcid.org/0000-0002-1061-604X","observed_orcids":["https://orcid.org/0000-0002-1061-604X"]},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Wang, Zezhong","raw_affiliation_strings":["The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100352418","display_name":"Xian Wu","orcid":"https://orcid.org/0000-0003-1118-9710","observed_orcids":["https://orcid.org/0000-0003-1118-9710","https://orcid.org/0000-0003-0840-5857"]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wu, Xian","raw_affiliation_strings":["Jarvis Lab , Tencent"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jarvis Lab , Tencent","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051649145","display_name":"Yefeng Zheng","orcid":"https://orcid.org/0000-0003-2195-2847","observed_orcids":["https://orcid.org/0000-0003-2195-2847"]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zheng, Yefeng","raw_affiliation_strings":["Jarvis Lab , Tencent"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jarvis Lab , Tencent","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5008208316","display_name":"Kam\u2010Fai Wong","orcid":"https://orcid.org/0000-0002-9427-5659","observed_orcids":["https://orcid.org/0000-0002-9427-5659"]},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Wong, Kam-Fai","raw_affiliation_strings":["The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":3,"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/T12031","display_name":"Speech and dialogue systems","score":1.0,"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"}},"topics":[{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":1.0,"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"}},{"id":"https://openalex.org/T12128","display_name":"AI in Service Interactions","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"}},{"id":"https://openalex.org/T11902","display_name":"Intelligent Tutoring Systems and Adaptive 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/dialogue-policy-learning","display_name":"dialogue policy learning","score":0.9950000047683716},{"id":"https://openalex.org/keywords/text-to-text-transformers","display_name":"text-to-text transformers","score":0.6690000295639038},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"reinforcement learning","score":0.7720000147819519},{"id":"https://openalex.org/keywords/reward-shaping","display_name":"reward shaping","score":0.9980000257492065},{"id":"https://openalex.org/keywords/dialogue-modeling","display_name":"dialogue modeling","score":0.6000000238418579},{"id":"https://openalex.org/keywords/response-diversity","display_name":"response diversity","score":0.8849999904632568},{"id":"https://openalex.org/keywords/edge-cases","display_name":"edge cases","score":0.6890000104904175}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.894307017326355},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8096970319747925},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7009867429733276},{"id":"https://openalex.org/C173853756","wikidata":"https://www.wikidata.org/wiki/Q86915","display_name":"Dialog box","level":2,"score":0.6872473955154419},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6208412051200867},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5581552982330322},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4122960567474365},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.07419785857200623},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2309.00230","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2309.00230","pdf_url":"https://arxiv.org/pdf/2309.00230","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2309.00230","doi":"https://doi.org/10.48550/arxiv.2309.00230","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2309.00230","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":"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2309.00230","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2309.00230","pdf_url":"https://arxiv.org/pdf/2309.00230","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4386437460.pdf","grobid_xml":"https://content.openalex.org/works/W4386437460.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4377232764","https://openalex.org/W4389010472","https://openalex.org/W7131788838","https://openalex.org/W7131780867","https://openalex.org/W4391305545","https://openalex.org/W4360891910","https://openalex.org/W7162283288","https://openalex.org/W4285606547","https://openalex.org/W2973825034","https://openalex.org/W2997108628"],"abstract_inverted_index":{"Dialogue":[0],"policy":[1,109],"learning":[2,150],"(DPL)":[3],"is":[4,14,88],"a":[5,35,45,82,93,107,113,152],"crucial":[6],"component":[7],"of":[8,56,133,181],"dialogue":[9,100,118,160,199],"modelling.":[10],"Its":[11],"primary":[12],"role":[13],"to":[15,23,97,140,155,166,183],"determine":[16],"the":[17,25,54,122,131,137,158,164],"appropriate":[18],"abstract":[19],"response,":[20],"commonly":[21],"referred":[22],"as":[24,34,90,202],"\"dialogue":[26],"action\".":[27],"Traditional":[28],"DPL":[29],"methodologies":[30],"have":[31],"treated":[32],"this":[33],"sequential":[36],"decision":[37],"problem,":[38],"using":[39],"pre-defined":[40],"action":[41,119,126],"candidates":[42,50],"extracted":[43],"from":[44,168],"corpus.":[46],"However,":[47],"these":[48,78],"incomplete":[49],"can":[51],"significantly":[52],"limit":[53],"diversity":[55,132],"responses":[57,134],"and":[58,116,135,208],"pose":[59],"challenges":[60],"when":[61],"dealing":[62],"with":[63,151],"edge":[64,142],"cases,":[65],"which":[66,162],"are":[67],"scenarios":[68],"that":[69,110,191],"occur":[70],"only":[71],"at":[72],"extreme":[73],"operating":[74],"parameters.":[75],"To":[76],"address":[77],"limitations,":[79],"we":[80],"introduce":[81],"novel":[83],"framework,":[84],"JoTR.":[85],"This":[86,128],"framework":[87],"unique":[89],"it":[91],"leverages":[92],"text-to-text":[94],"Transformer-based":[95],"model":[96,165],"generate":[98],"flexible":[99],"actions.":[101],"Unlike":[102],"traditional":[103],"methods,":[104],"JoTR":[105,147,182,192],"formulates":[106],"word-level":[108,159],"allows":[111,163],"for":[112,124],"more":[114],"dynamic":[115],"adaptable":[117],"generation,":[120],"without":[121],"need":[123],"any":[125],"templates.":[127],"setting":[129],"enhances":[130],"improves":[136],"system's":[138],"ability":[139],"handle":[141],"cases":[143],"effectively.":[144],"In":[145],"addition,":[146],"employs":[148],"reinforcement":[149],"reward-shaping":[153],"mechanism":[154],"efficiently":[156],"finetune":[157],"policy,":[161],"learn":[167],"its":[169,172,185],"interactions,":[170],"improving":[171],"performance":[173,195],"over":[174],"time.":[175],"We":[176],"conducted":[177],"an":[178],"extensive":[179,188],"evaluation":[180,189],"assess":[184],"effectiveness.":[186],"Our":[187],"shows":[190],"achieves":[193],"state-of-the-art":[194],"on":[196],"two":[197],"benchmark":[198],"modelling":[200],"tasks,":[201],"assessed":[203],"by":[204],"both":[205],"user":[206],"simulators":[207],"human":[209],"evaluators.":[210]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":2}],"updated_date":"2026-11-01T11:40:01.658920","created_date":"2023-09-06T00:00:00"}