{"id":"https://openalex.org/W4416034779","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.696","title":"Logic-Thinker: Teaching Large Language Models to Think more Logically.","display_name":"Logic-Thinker: Teaching Large Language Models to Think more Logically.","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034779","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.696"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.696","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.696","pdf_url":"https://aclanthology.org/2025.findings-emnlp.696.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: EMNLP 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.696.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Chengyao Wen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chengyao Wen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083141785","display_name":"Qiang Cheng","orcid":"https://orcid.org/0000-0002-2442-8357"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qiang Cheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066444643","display_name":"S. Wang","orcid":"https://orcid.org/0000-0003-2697-5335"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shaofei Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101917237","display_name":"Zhizhen Liu","orcid":"https://orcid.org/0000-0002-4485-8688"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhizhen Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110051234","display_name":"Zhao Deng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Deng Zhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100322370","display_name":"Lei Liang","orcid":"https://orcid.org/0000-0002-5397-0284"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lei Liang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"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":0.15748632,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"12955","last_page":"12969"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12072","display_name":"Machine Learning and Algorithms","score":0.18050000071525574,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.18050000071525574,"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/T10028","display_name":"Topic Modeling","score":0.10209999978542328,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.06689999997615814,"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/language-model","display_name":"Language model","score":0.2985000014305115},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.29510000348091125},{"id":"https://openalex.org/keywords/modeling-language","display_name":"Modeling language","score":0.2667999863624573},{"id":"https://openalex.org/keywords/on-language","display_name":"On Language","score":0.263700008392334}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5357000231742859},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.49799999594688416},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3260999917984009},{"id":"https://openalex.org/C145420912","wikidata":"https://www.wikidata.org/wiki/Q853077","display_name":"Mathematics education","level":1,"score":0.32030001282691956},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.3197999894618988},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3176000118255615},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2985000014305115},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.29510000348091125},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.2667999863624573},{"id":"https://openalex.org/C2779313563","wikidata":"https://www.wikidata.org/wiki/Q17072565","display_name":"On Language","level":2,"score":0.263700008392334},{"id":"https://openalex.org/C129353971","wikidata":"https://www.wikidata.org/wiki/Q5156949","display_name":"Comprehension approach","level":3,"score":0.26179999113082886}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.696","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.696","pdf_url":"https://aclanthology.org/2025.findings-emnlp.696.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: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.696","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.696","pdf_url":"https://aclanthology.org/2025.findings-emnlp.696.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: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034779.pdf","grobid_xml":"https://content.openalex.org/works/W4416034779.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"Large":[1],"Reasoning":[2],"Models":[3],"(LRMs)":[4],"have":[5],"demonstrated":[6],"the":[7,130],"ability":[8],"to":[9,43,69,87],"generate":[10],"long":[11],"chains":[12,83],"of":[13,84,120,134],"thought":[14],"(LongCoT)":[15],"before":[16],"arriving":[17],"at":[18],"a":[19,61],"final":[20],"conclusion.Despite":[21],"remarkable":[22],"breakthroughs":[23],"in":[24,98],"complex":[25],"reasoning":[26,54,63,100,113,132,142],"capabilities,":[27],"LongCoT":[28],"still":[29],"faces":[30],"challenges":[31],"such":[32,144],"as":[33,88,136,145],"redundancy":[34],"and":[35,51,73,81,147],"logical":[36,53,99],"incoherence.To":[37],"address":[38],"these":[39],"issues,":[40],"we":[41,58],"aim":[42],"equip":[44],"large":[45],"language":[46],"models":[47,107],"(LLMs)":[48],"with":[49,104],"rigorous":[50,82],"concise":[52,80],"capabilities.In":[55],"this":[56],"work,":[57],"propose":[59],"Logic-Thinker,":[60],"neural-symbolic":[62],"framework":[64],"that":[65,93],"employs":[66],"symbolic":[67],"solvers":[68],"precisely":[70],"solve":[71],"problems":[72],"transforms":[74],"their":[75],"internal":[76],"solving":[77],"processes":[78],"into":[79],"thought,":[85],"referred":[86],"Thinker-CoT.Our":[89],"experimental":[90],"results":[91],"demonstrate":[92],"Logic-Thinker":[94],"achieves":[95],"state-of-the-art":[96],"performance":[97,139],"problems.Additionally,":[101],"LLMs":[102],"fine-tuned":[103],"ThinkerCoT":[105,128],"outperform":[106],"distilled":[108],"from":[109],"QwQ32B":[110],"on":[111,141],"logic":[112],"tasks,":[114],"achieving":[115],"an":[116],"overall":[117],"accuracy":[118],"improvement":[119],"3.6%":[121],"while":[122],"reducing":[123],"token":[124],"output":[125],"by":[126,138],"73%-91%.Furthermore,":[127],"enhances":[129],"comprehensive":[131],"capabilities":[133],"LLMs,":[135],"evidenced":[137],"improvements":[140],"benchmarks":[143],"GPQA":[146],"AIME.":[148]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-11-08T00:00:00"}