{"id":"https://openalex.org/W7166878344","doi":"https://doi.org/10.18653/v1/2026.acl-long.613","title":"Android Coach: Improve Online Agentic Training Efficiency with Single State Multiple Actions","display_name":"Android Coach: Improve Online Agentic Training Efficiency with Single State Multiple Actions","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166878344","doi":"https://doi.org/10.18653/v1/2026.acl-long.613"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.613","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.613","pdf_url":"https://aclanthology.org/2026.acl-long.613.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.613.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139797406","display_name":"Guo Gan","orcid":null},"institutions":[{"id":"https://openalex.org/I19268510","display_name":"Qualcomm (United Kingdom)","ror":"https://ror.org/04d3djg48","country_code":"GB","type":"company","lineage":["https://openalex.org/I19268510","https://openalex.org/I4210087596"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Guo Gan","raw_affiliation_strings":["Zhejiang University Q Qualcomm AI Research initiative of Qualcomm Technologies, Inc. experi- ence"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University Q Qualcomm AI Research initiative of Qualcomm Technologies, Inc. experi- ence","institution_ids":["https://openalex.org/I19268510"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139748402","display_name":"Yuxuan Ding","orcid":null},"institutions":[{"id":"https://openalex.org/I19268510","display_name":"Qualcomm (United Kingdom)","ror":"https://ror.org/04d3djg48","country_code":"GB","type":"company","lineage":["https://openalex.org/I19268510","https://openalex.org/I4210087596"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yuxuan Ding","raw_affiliation_strings":["Zhejiang University Q Qualcomm AI Research initiative of Qualcomm Technologies, Inc. experi- ence"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University Q Qualcomm AI Research initiative of Qualcomm Technologies, Inc. experi- ence","institution_ids":["https://openalex.org/I19268510"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139755832","display_name":"Cong Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cong Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139783256","display_name":"Yuwei Ren","orcid":null},"institutions":[{"id":"https://openalex.org/I19268510","display_name":"Qualcomm (United Kingdom)","ror":"https://ror.org/04d3djg48","country_code":"GB","type":"company","lineage":["https://openalex.org/I19268510","https://openalex.org/I4210087596"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yuwei Ren","raw_affiliation_strings":["Zhejiang University Q Qualcomm AI Research initiative of Qualcomm Technologies, Inc. experi- ence"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University Q Qualcomm AI Research initiative of Qualcomm Technologies, Inc. experi- ence","institution_ids":["https://openalex.org/I19268510"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139760377","display_name":"Yin Huang","orcid":null},"institutions":[{"id":"https://openalex.org/I19268510","display_name":"Qualcomm (United Kingdom)","ror":"https://ror.org/04d3djg48","country_code":"GB","type":"company","lineage":["https://openalex.org/I19268510","https://openalex.org/I4210087596"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Yin Huang","raw_affiliation_strings":["Zhejiang University Q Qualcomm AI Research initiative of Qualcomm Technologies, Inc. experi- ence"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University Q Qualcomm AI Research initiative of Qualcomm Technologies, Inc. experi- ence","institution_ids":["https://openalex.org/I19268510"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5139757258","display_name":"Hong Zhou","orcid":null},"institutions":[{"id":"https://openalex.org/I19268510","display_name":"Qualcomm (United Kingdom)","ror":"https://ror.org/04d3djg48","country_code":"GB","type":"company","lineage":["https://openalex.org/I19268510","https://openalex.org/I4210087596"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Hong Zhou","raw_affiliation_strings":["Zhejiang University Q Qualcomm AI Research initiative of Qualcomm Technologies, Inc. experi- ence"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University Q Qualcomm AI Research initiative of Qualcomm Technologies, Inc. experi- ence","institution_ids":["https://openalex.org/I19268510"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.86620917,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"13407","last_page":"13428"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10456","display_name":"Multi-Agent Systems and Negotiation","score":0.13490000367164612,"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/T10456","display_name":"Multi-Agent Systems and Negotiation","score":0.13490000367164612,"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/T12203","display_name":"Mobile Agent-Based Network Management","score":0.09440000355243683,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T11574","display_name":"Artificial Intelligence in Games","score":0.05829999968409538,"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/training","display_name":"Training (meteorology)","score":0.47850000858306885},{"id":"https://openalex.org/keywords/android","display_name":"Android (operating system)","score":0.35499998927116394},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.3456000089645386},{"id":"https://openalex.org/keywords/mobile-device","display_name":"Mobile device","score":0.33079999685287476},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.2777000069618225}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6326000094413757},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.47850000858306885},{"id":"https://openalex.org/C557433098","wikidata":"https://www.wikidata.org/wiki/Q94","display_name":"Android (operating system)","level":2,"score":0.35499998927116394},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.3456000089645386},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.33079999685287476},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3303999900817871},{"id":"https://openalex.org/C49774154","wikidata":"https://www.wikidata.org/wiki/Q131765","display_name":"Multimedia","level":1,"score":0.2928999960422516},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2912999987602234},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.2777000069618225},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.2597000002861023}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.613","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.613","pdf_url":"https://aclanthology.org/2026.acl-long.613.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.613","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.613","pdf_url":"https://aclanthology.org/2026.acl-long.613.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5949402452","display_name":null,"funder_award_id":"2025C02104","funder_id":"https://openalex.org/F4320338469","funder_display_name":"Science and Technology Program of Zhejiang Province"}],"funders":[{"id":"https://openalex.org/F4320338469","display_name":"Science and Technology Program of Zhejiang Province","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166878344.pdf","grobid_xml":"https://content.openalex.org/works/W7166878344.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Online":[0],"reinforcement":[1],"learning":[2,111],"(RL)":[3],"serves":[4,121],"as":[5,122],"an":[6],"effective":[7],"method":[8],"for":[9,98],"enhancing":[10],"Android":[11],"agents.However,":[12],"guiding":[13],"agents":[14],"to":[15,24,84,92],"learn":[16],"through":[17],"online":[18,60,101,110],"interaction":[19],"is":[20],"prohibitively":[21],"expensive":[22],"due":[23],"the":[25,31,45,53,81,90,119,140,146],"high":[26],"latency":[27],"of":[28,34,150],"emulators":[29],"and":[30,94,132,148,156,163,179],"sample":[32,93],"inefficiency":[33],"existing":[35],"RL":[36],"algorithms.We":[37],"identify":[38],"a":[39,76,99,112,123,128,134],"fundamental":[40],"limitation":[41],"in":[42],"current":[43],"approaches:":[44],"Single":[46,48,85,173,175],"State":[47,86,174],"Action":[49,176],"paradigm,":[50],"which":[51],"updates":[52],"policy":[54],"with":[55],"one-to-one":[56],"state-action":[57],"pairs":[58],"from":[59],"one-way":[61],"rollouts,":[62],"without":[63,105],"fully":[64],"exploring":[65],"each":[66],"costly":[67],"emulator":[68,107],"state.In":[69],"this":[70,104],"paper,":[71],"we":[72,126],"propose":[73],"ANDROID":[74,151],"COACH,":[75],"novel":[77],"framework":[78],"that":[79,114],"shifts":[80],"training":[82,170],"paradigm":[83],"Multiple":[87],"Actions,":[88],"allowing":[89],"agent":[91],"utilize":[95],"multiple":[96],"actions":[97],"single":[100],"state.We":[102],"enable":[103],"additional":[106],"overhead":[108],"by":[109],"critic":[113,120,142],"estimates":[115],"action":[116],"values.To":[117],"ensure":[118],"reliable":[124],"coach,":[125],"integrate":[127],"process":[129],"reward":[130],"model":[131],"introduce":[133],"group-wise":[135],"advantage":[136],"estimator":[137],"based":[138],"on":[139,161],"averaged":[141],"outputs.Extensive":[143],"experiments":[144],"demonstrate":[145],"effectiveness":[147],"efficiency":[149,171],"COACH:":[152],"it":[153],"achieves":[154],"7.5%":[155],"8.3%":[157],"success":[158,183],"rate":[159],"improvements":[160],"Android-Lab":[162],"AndroidWorld":[164],"over":[165],"UI-TARS-1.5-7B,and":[166],"attains":[167],"1.4\u00d7":[168],"higher":[169],"than":[172],"methods":[177],"PPO":[178],"GRPO":[180],"at":[181],"matched":[182],"rates.":[184]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}