{"id":"https://openalex.org/W7166803363","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1530","title":"DARL: Encouraging Diverse Answers for General Reasoning without Verifiers","display_name":"DARL: Encouraging Diverse Answers for General Reasoning without Verifiers","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166803363","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1530"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1530","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1530","pdf_url":"https://aclanthology.org/2026.findings-acl.1530.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.1530.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5007564034","display_name":"Chongxuan Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chongxuan Huang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139776727","display_name":"Lei Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lei Lin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139817758","display_name":"Xiaodong Shi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaodong Shi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139715137","display_name":"Wenping Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wenping Hu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139760894","display_name":"Ruiming Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ruiming Tang","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":0.7928769,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"30649","last_page":"30665"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.32190001010894775,"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":0.32190001010894775,"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":0.14550000429153442,"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":0.1054999977350235,"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/class","display_name":"Class (philosophy)","score":0.31380000710487366},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.3122999966144562},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.30070000886917114},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.26269999146461487},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.2524999976158142}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4327999949455261},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42500001192092896},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.31380000710487366},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.3122999966144562},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.30070000886917114},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.27410000562667847},{"id":"https://openalex.org/C111472728","wikidata":"https://www.wikidata.org/wiki/Q9471","display_name":"Epistemology","level":1,"score":0.2712000012397766},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.26269999146461487},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.2524999976158142},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2483000010251999}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1530","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1530","pdf_url":"https://aclanthology.org/2026.findings-acl.1530.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.1530","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1530","pdf_url":"https://aclanthology.org/2026.findings-acl.1530.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":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166803363.pdf","grobid_xml":"https://content.openalex.org/works/W7166803363.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"Learning":[1],"with":[2,121,127],"Verifiable":[3],"Rewards":[4],"(RLVR)":[5],"has":[6],"demonstrated":[7],"promising":[8],"gains":[9,156],"in":[10,80,148,174],"enhancing":[11],"the":[12,69,105,116],"reasoning":[13,149,162,177],"capabilities":[14],"of":[15,56,107,157],"large":[16],"language":[17],"models.However,":[18],"its":[19,26,172],"dependence":[20],"on":[21,45,142,160,167],"domain-specific":[22],"verifiers":[23],"significantly":[24],"restricts":[25],"applicability":[27],"to":[28,40,62,64,72],"open":[29],"and":[30,48,133,164,179],"general":[31,41,129,169],"domains.Recent":[32],"efforts":[33],"such":[34,83],"as":[35,84],"RLPR":[36],"have":[37],"extended":[38],"RLVR":[39],"domains,":[42],"enabling":[43],"training":[44],"broader":[46],"datasets":[47],"achieving":[49,154],"improvements":[50,147],"over":[51],"RLVR.However,":[52],"a":[53,96,111],"notable":[54],"limitation":[55,76],"these":[57],"methods":[58,132],"is":[59,77,124],"their":[60],"tendency":[61],"overfit":[63],"reference":[65,117],"answers,":[66],"which":[67],"constrains":[68],"model's":[70],"ability":[71],"generate":[73],"diverse":[74,108],"outputs.This":[75],"particularly":[78],"pronounced":[79],"open-ended":[81],"tasks":[82],"writing,":[85],"where":[86],"multiple":[87],"plausible":[88],"answers":[89,109],"exist.To":[90],"address":[91],"this,":[92],"we":[93],"propose":[94],"DARL,":[95],"simple":[97],"yet":[98],"effective":[99],"reinforcement":[100,130],"learning":[101,131],"framework":[102,123],"that":[103],"encourages":[104],"generation":[106],"within":[110],"controlled":[112],"deviation":[113],"range":[114],"from":[115],"while":[118],"preserving":[119],"alignment":[120],"it.Our":[122],"fully":[125],"compatible":[126],"existing":[128],"can":[134],"be":[135],"seamlessly":[136],"integrated":[137],"without":[138],"additional":[139],"verifiers.Extensive":[140],"experiments":[141],"thirteen":[143],"benchmarks":[144,163],"demonstrate":[145],"overall":[146],"performance.Notably,":[150],"DARL":[151],"surpasses":[152],"RLPR,":[153],"average":[155],"1.3":[158],"points":[159,166],"six":[161],"9.5":[165],"seven":[168],"benchmarks,":[170],"highlighting":[171],"effectiveness":[173],"improving":[175],"both":[176],"accuracy":[178],"output":[180],"diversity.":[181]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}