{"id":"https://openalex.org/W7140090787","doi":"https://doi.org/10.18653/v1/2026.findings-eacl.3","title":"WorkForceAgent-R1: Incentivizing Reasoning Capability in LLM-based Web Agents via Reinforcement Learning","display_name":"WorkForceAgent-R1: Incentivizing Reasoning Capability in LLM-based Web Agents via Reinforcement Learning","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7140090787","doi":"https://doi.org/10.18653/v1/2026.findings-eacl.3"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-eacl.3","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.3","pdf_url":"https://aclanthology.org/2026.findings-eacl.3.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":"Findings of the Association for Computational Linguistics: EACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-eacl.3.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130357694","display_name":"Yuchen Zhuang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuchen Zhuang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130386805","display_name":"Di Jin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Di Jin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130395015","display_name":"Jiaao Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiaao Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130357620","display_name":"Wenqi Shi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wenqi Shi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130341482","display_name":"Hanrui Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hanrui Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130375563","display_name":"Chao Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chao Zhang","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":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"34","last_page":"49"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10456","display_name":"Multi-Agent Systems and Negotiation","score":0.5310999751091003,"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":0.5310999751091003,"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":0.03009999915957451,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.0284000001847744,"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":0.5594000220298767},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.30649998784065247},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.2957000136375427},{"id":"https://openalex.org/keywords/reinforcement","display_name":"Reinforcement","score":0.2777000069618225},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.2721000015735626},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.2694999873638153}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.63919997215271},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5594000220298767},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42559999227523804},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.32420000433921814},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.30649998784065247},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.2957000136375427},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.2777000069618225},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2721000015735626},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.26919999718666077},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.26330000162124634},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.25519999861717224},{"id":"https://openalex.org/C41550386","wikidata":"https://www.wikidata.org/wiki/Q529909","display_name":"Multi-agent system","level":2,"score":0.2531000077724457}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-eacl.3","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.3","pdf_url":"https://aclanthology.org/2026.findings-eacl.3.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":"Findings of the Association for Computational Linguistics: EACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-eacl.3","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-eacl.3","pdf_url":"https://aclanthology.org/2026.findings-eacl.3.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":"Findings of the Association for Computational Linguistics: EACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7140090787.pdf","grobid_xml":"https://content.openalex.org/works/W7140090787.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)-empowered":[3],"web":[4,10,17,42,53,73,131],"agents":[5,18,127],"enables":[6],"automating":[7],"complex,":[8],"realtime":[9],"navigation":[11,74,132],"tasks":[12],"in":[13,129],"enterprise":[14],"environments.However,":[15],"existing":[16],"relying":[19],"on":[20,108],"supervised":[21],"fine-tuning":[22],"(SFT)":[23],"often":[24],"struggle":[25],"with":[26],"generalization":[27],"and":[28,69,88],"robustness":[29],"due":[30],"to":[31,65,85,94,124],"insufficient":[32],"reasoning":[33,68,99],"capabilities":[34],"when":[35],"handling":[36],"the":[37,109],"inherently":[38],"dynamic":[39],"nature":[40],"of":[41,90],"interactions.In":[43],"this":[44],"study,":[45],"we":[46],"introduce":[47],"WorkForceAgent-R1":[48,93,114],"1":[49],",":[50],"an":[51],"LLM-based":[52,126],"agent":[54],"trained":[55],"using":[56],"a":[57,77],"rule-based":[58],"R1-style":[59],"reinforcement":[60],"learning":[61],"framework":[62],"designed":[63],"explicitly":[64],"enhance":[66],"single-step":[67],"planning":[70],"for":[71],"business-oriented":[72],"tasks.We":[75],"employ":[76],"structured":[78],"reward":[79],"function":[80],"that":[81,113],"evaluates":[82],"both":[83],"adherence":[84],"output":[86],"formats":[87],"correctness":[89],"actions,":[91],"enabling":[92],"implicitly":[95],"learn":[96],"robust":[97],"intermediate":[98],"without":[100],"explicit":[101],"annotations":[102],"or":[103],"extensive":[104],"expert":[105],"demonstrations.Extensive":[106],"experiments":[107],"WorkArena":[110],"benchmark":[111],"demonstrate":[112],"substantially":[115],"outperforms":[116],"SFT":[117],"baselines":[118],"by":[119],"10.26-16.59%,achieving":[120],"competitive":[121],"performance":[122],"relative":[123],"proprietary":[125],"(gpt-4o)":[128],"workplace-oriented":[130],"tasks.":[133]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-03-24T00:00:00"}