{"id":"https://openalex.org/W7166850182","doi":"https://doi.org/10.18653/v1/2026.findings-acl.198","title":"TemplateRL: Structured Template-Guided Reinforcement Learning for LLM Reasoning","display_name":"TemplateRL: Structured Template-Guided Reinforcement Learning for LLM Reasoning","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166850182","doi":"https://doi.org/10.18653/v1/2026.findings-acl.198"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.198","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.198","pdf_url":"https://aclanthology.org/2026.findings-acl.198.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: ACL 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-acl.198.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139798608","display_name":"Jinyang Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinyang Wu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064973064","display_name":"Chonghua Liao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chonghua Liao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139750374","display_name":"Mingkuan Feng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mingkuan Feng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139728445","display_name":"Shuai Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shuai Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139852042","display_name":"Zhengqi Wen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhengqi Wen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139807805","display_name":"Haoran Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Haoran Luo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139848410","display_name":"Ling Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ling Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139824228","display_name":"Huazhe Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huazhe Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139733429","display_name":"Jianhua Tao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jianhua Tao","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":0.83971699,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4050","last_page":"4080"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.15639999508857727,"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/T10028","display_name":"Topic Modeling","score":0.15639999508857727,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.11810000240802765,"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":0.05719999969005585,"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":0.38100001215934753},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.33959999680519104},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.2939999997615814},{"id":"https://openalex.org/keywords/case-based-reasoning","display_name":"Case-based reasoning","score":0.2865000069141388},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.2858999967575073},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.26600000262260437}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5356000065803528},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45559999346733093},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.38100001215934753},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.33959999680519104},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2939999997615814},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.2865000069141388},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.2858999967575073},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.26600000262260437},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.26010000705718994},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.25850000977516174},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.25380000472068787},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2533999979496002}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.198","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.198","pdf_url":"https://aclanthology.org/2026.findings-acl.198.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: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.198","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.198","pdf_url":"https://aclanthology.org/2026.findings-acl.198.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: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5784019134","display_name":null,"funder_award_id":"U2436210","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G7003438179","display_name":null,"funder_award_id":"62322120","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166850182.pdf","grobid_xml":"https://content.openalex.org/works/W7166850182.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1],"(RL)":[2],"has":[3],"emerged":[4],"as":[5],"an":[6],"effective":[7],"paradigm":[8],"for":[9,163],"enhancing":[10,114],"model":[11],"reasoning.However,":[12],"existing":[13],"RL":[14,47,79,115],"methods":[15],"like":[16],"GRPO":[17,141],"typically":[18],"rely":[19],"on":[20,66,144,148,153],"unstructured":[21],"self-sampling":[22],"to":[23,33,84],"fit":[24],"scalar":[25],"rewards,":[26],"often":[27],"producing":[28],"inefficient":[29],"rollouts":[30],"that":[31,49,138],"fail":[32],"capture":[34],"transferable":[35],"problem-solving":[36,61],"strategies.To":[37],"address":[38],"this":[39,74],"limitation,":[40],"we":[41],"propose":[42],"TemplateRL,":[43],"a":[44,60,67],"structured":[45,76],"template-guided":[46],"framework":[48],"augments":[50],"policy":[51,105],"optimization":[52],"with":[53,86,150],"explicit":[54,119],"template":[55,62,88,120],"guidance.Our":[56],"approach":[57],"first":[58],"constructs":[59],"library":[63,121],"via":[64],"MCTS":[65],"small":[68],"seed":[69],"set,":[70],"then":[71],"seamlessly":[72],"integrates":[73],"high-level":[75],"guidance":[77],"into":[78],"training.By":[80],"guiding":[81],"rollout":[82],"generation":[83],"align":[85],"proven":[87],"structures,":[89],"TemplateRL":[90,139],"significantly":[91],"improves":[92],"high-quality":[93],"trajectory":[94],"hit":[95],"rates":[96],"while":[97],"reducing":[98],"ineffective":[99],"exploration.This":[100],"structure-guided":[101],"design":[102],"steers":[103],"the":[104,118],"toward":[106],"validated":[107],"strategic":[108],"patterns,":[109],"stabilizing":[110],"training":[111,133],"dynamics,":[112],"and":[113,125,134,146,156],"sampling":[116],"efficiency.Notably,":[117],"is":[122],"interpretable,":[123],"editable,":[124],"supports":[126],"online":[127],"updates-enabling":[128],"continuous":[129],"updates":[130],"during":[131],"both":[132],"inference.Extensive":[135],"experiments":[136],"demonstrate":[137],"outperforms":[140],"by":[142],"99%":[143],"AIME":[145],"41%":[147],"AMC,":[149],"superior":[151],"stability":[152],"weak":[154],"models":[155],"remarkable":[157],"cross-domain":[158],"generalization,":[159],"highlighting":[160],"its":[161],"potential":[162],"broader":[164],"tasks.":[165]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}