{"id":"https://openalex.org/W7166881334","doi":"https://doi.org/10.18653/v1/2026.acl-long.1624","title":"SOAR: Supervision from Observation for Agentic Reinforcement Learning","display_name":"SOAR: Supervision from Observation for Agentic Reinforcement Learning","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166881334","doi":"https://doi.org/10.18653/v1/2026.acl-long.1624"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.1624","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1624","pdf_url":"https://aclanthology.org/2026.acl-long.1624.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.1624.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139762835","display_name":"Meng Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Meng Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139778197","display_name":"Lei Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lei Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139791005","display_name":"Xiting Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiting Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139767178","display_name":"Yi Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yi Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139774692","display_name":"Zheng Wei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139829451","display_name":"Brucebian","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Brucebian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5081888630","display_name":"Li Zang","orcid":"https://orcid.org/0000-0002-3183-8899"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zang Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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":0.86974678,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"35175","last_page":"35197"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.7781999707221985,"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":0.7781999707221985,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.018400000408291817,"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":0.01810000091791153,"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","display_name":"Reinforcement","score":0.35740000009536743},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.33709999918937683},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.3050000071525574},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.302700012922287},{"id":"https://openalex.org/keywords/experiential-learning","display_name":"Experiential learning","score":0.2849000096321106},{"id":"https://openalex.org/keywords/agency","display_name":"Agency (philosophy)","score":0.265500009059906}],"concepts":[{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.5932999849319458},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.40049999952316284},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.38530001044273376},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.35740000009536743},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.33709999918937683},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.3050000071525574},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.302700012922287},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.29019999504089355},{"id":"https://openalex.org/C37228920","wikidata":"https://www.wikidata.org/wiki/Q1307600","display_name":"Experiential learning","level":2,"score":0.2849000096321106},{"id":"https://openalex.org/C108170787","wikidata":"https://www.wikidata.org/wiki/Q3951828","display_name":"Agency (philosophy)","level":2,"score":0.265500009059906},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.26330000162124634},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.25440001487731934},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2531999945640564},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.25209999084472656}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.1624","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1624","pdf_url":"https://aclanthology.org/2026.acl-long.1624.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.1624","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.1624","pdf_url":"https://aclanthology.org/2026.acl-long.1624.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/G3248486410","display_name":null,"funder_award_id":"U2436209","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G4128466783","display_name":null,"funder_award_id":"92470205","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G6750762792","display_name":null,"funder_award_id":"62476279","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320316083","display_name":"Tencent","ror":"https://ror.org/00hhjss72"},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320322499","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92"},{"id":"https://openalex.org/F4320335787","display_name":"Fundamental Research Funds for the Central Universities","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166881334.pdf","grobid_xml":"https://content.openalex.org/works/W7166881334.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Agentic":[0,68],"reinforcement":[1],"learning":[2],"enables":[3],"large":[4],"language":[5],"models":[6],"to":[7,75,79,89,125],"solve":[8],"long-horizon":[9],"tasks":[10,130],"by":[11],"interacting":[12],"with":[13],"the":[14,53,80,87],"environment":[15,38,54,100],"and":[16,102,112,131,140],"internalizing":[17],"tool-use":[18],"behavior":[19],"into":[20],"their":[21],"reasoning.Prior":[22],"work":[23],"assigns":[24,72],"supervision":[25],"primarily":[26],"based":[27],"on":[28,127,133],"outcome":[29],"rewards":[30],"or":[31],"external":[32],"reward":[33],"models,":[34],"but":[35],"largely":[36],"ignores":[37],"observations,":[39],"a":[40],"critical":[41],"source":[42],"of":[43,83,93,105,123],"learning.Consequently,":[44],"agents":[45],"may":[46],"identify":[47],"successful":[48],"actions":[49],"without":[50],"understanding":[51],"how":[52],"responds,":[55],"producing":[56],"suboptimal":[57],"policies.To":[58],"address":[59],"this,":[60],"we":[61],"propose":[62],"SOAR":[63,117],"(Supervision":[64],"from":[65,91],"Observation":[66],"for":[67],"Reinforcement":[69],"Learning),":[70],"which":[71],"positive":[73],"advantages":[74],"observation":[76],"tokens":[77],"proportional":[78],"negative":[81],"entropy":[82],"preceding":[84],"actions.This":[85],"encourages":[86],"agent":[88],"learn":[90],"outcomes":[92],"confident":[94],"actions,":[95],"grounding":[96],"policy":[97],"updates":[98],"in":[99],"dynamics":[101],"improving":[103],"anticipation":[104],"tool-call":[106],"consequences.Empirical":[107],"results":[108],"across":[109],"three":[110],"domains":[111],"13":[113],"benchmarks":[114],"show":[115],"that":[116],"consistently":[118],"improves":[119],"performance,":[120],"yielding":[121],"gains":[122],"up":[124],"7.0%":[126],"general":[128],"reasoning":[129],"16.9%":[132],"deep":[134],"research":[135],"tasks,":[136],"while":[137],"reducing":[138],"erroneous":[139],"inefficient":[141],"tool":[142],"usage.":[143]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}