{"id":"https://openalex.org/W7166824104","doi":"https://doi.org/10.18653/v1/2026.acl-long.649","title":"Discovery and Reinforcement of Tool-Integrated Reasoning Chains via Rollout Trees","display_name":"Discovery and Reinforcement of Tool-Integrated Reasoning Chains via Rollout Trees","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166824104","doi":"https://doi.org/10.18653/v1/2026.acl-long.649"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.649","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.649","pdf_url":"https://aclanthology.org/2026.acl-long.649.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.649.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139826050","display_name":"Kun Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kun Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043757700","display_name":"Zenan Xu","orcid":"https://orcid.org/0000-0003-1662-0063"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zenan Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139837673","display_name":"Junan Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Junan Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139817725","display_name":"Zengrui Jin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zengrui Jin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139780392","display_name":"Jinghao Deng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinghao Deng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122481409","display_name":"Zexuan Qiu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zexuan Qiu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139814021","display_name":"Bo Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bo Zhou","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":0.81398133,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"14265","last_page":"14283"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10906","display_name":"AI-based Problem Solving and Planning","score":0.12559999525547028,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.12559999525547028,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.10369999706745148,"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/T11010","display_name":"Logic, Reasoning, and Knowledge","score":0.09830000251531601,"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":0.38089999556541443},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.3625999987125397},{"id":"https://openalex.org/keywords/tree","display_name":"Tree (set theory)","score":0.31929999589920044},{"id":"https://openalex.org/keywords/variety","display_name":"Variety (cybernetics)","score":0.28349998593330383},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.2799000144004822}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5490999817848206},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43540000915527344},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.38089999556541443},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.3625999987125397},{"id":"https://openalex.org/C113174947","wikidata":"https://www.wikidata.org/wiki/Q2859736","display_name":"Tree (set theory)","level":2,"score":0.31929999589920044},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.28349998593330383},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.2799000144004822},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2676999866962433},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.26100000739097595},{"id":"https://openalex.org/C113336015","wikidata":"https://www.wikidata.org/wiki/Q574010","display_name":"Complete information","level":2,"score":0.24799999594688416}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.649","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.649","pdf_url":"https://aclanthology.org/2026.acl-long.649.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.649","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.649","pdf_url":"https://aclanthology.org/2026.acl-long.649.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":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166824104.pdf","grobid_xml":"https://content.openalex.org/works/W7166824104.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Tool-Integrated":[0,57],"Reasoning":[1,58],"has":[2],"emerged":[3],"as":[4],"a":[5,63,101],"key":[6],"paradigm":[7],"to":[8,29,87,97,116],"augment":[9],"Large":[10],"Language":[11],"Models":[12],"(LLMs)":[13],"with":[14,145],"computational":[15],"capabilities,":[16],"yet":[17],"integrating":[18,39],"tool-use":[19,40,70,90],"into":[20],"long":[21,72,146],"Chainof-Thought":[22],"(long":[23],"CoT)":[24],"remains":[25],"underexplored,":[26],"largely":[27],"due":[28],"the":[30,36,43,117],"scarcity":[31],"of":[32,38,56],"training":[33,86],"data":[34],"and":[35,107,132],"challenge":[37],"without":[41,75],"compromising":[42],"model's":[44],"intrinsic":[45],"longchain":[46],"reasoning.In":[47],"this":[48],"paper,":[49],"we":[50],"introduce":[51],"DART":[52,136],"(Discovery":[53],"And":[54],"Reinforcement":[55],"Chains":[59],"via":[60],"Rollout":[61],"Trees),":[62],"reinforcement":[64],"learning":[65],"framework":[66],"that":[67,135],"enables":[68],"spontaneous":[69],"during":[71,85,124],"CoT":[73,147],"reasoning":[74],"additional":[76],"human":[77],"annotation.DART":[78],"operates":[79],"by":[80],"constructing":[81],"dynamic":[82],"rollout":[83],"trees":[84],"discover":[88],"valid":[89],"opportunities,":[91],"branching":[92],"out":[93],"at":[94],"promising":[95],"positions":[96],"explore":[98],"tool-integrated":[99],"trajectories.Subsequently,":[100],"tree-based":[102],"process":[103],"advantage":[104],"estimation":[105],"identifies":[106],"credits":[108],"specific":[109],"sub-trajectories":[110],"where":[111],"tool":[112,143],"invocation":[113],"positively":[114],"contributes":[115],"solution,":[118],"effectively":[119],"reinforcing":[120],"these":[121],"beneficial":[122],"behaviors":[123],"training.Extensive":[125],"experiments":[126],"on":[127],"challenging":[128],"benchmarks":[129],"like":[130],"AIME":[131],"GPQA-Diamond":[133],"demonstrate":[134],"significantly":[137],"outperforms":[138],"existing":[139],"methods,":[140],"successfully":[141],"harmonizing":[142],"execution":[144],"reasoning.":[148]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}