{"id":"https://openalex.org/W7138880913","doi":"https://doi.org/10.48550/arxiv.2603.17775","title":"CoVerRL: Breaking the Consensus Trap in Label-Free Reasoning via Generator-Verifier Co-Evolution","display_name":"CoVerRL: Breaking the Consensus Trap in Label-Free Reasoning via Generator-Verifier Co-Evolution","publication_year":2026,"publication_date":"2026-03-18","ids":{"openalex":"https://openalex.org/W7138880913","doi":"https://doi.org/10.48550/arxiv.2603.17775"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.17775","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17775","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.17775","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130059365","display_name":"Teng Pan","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Pan, Teng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129966668","display_name":"Yuchen Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Yuchen","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129906759","display_name":"Zixuan Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zixuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129761794","display_name":"Ruiqing Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Ruiqing","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130107725","display_name":"Gaiyang Han","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hou, Guiyang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zhang, Wenqi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Wenqi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130006836","display_name":"Weiming Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Weiming","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130176301","display_name":"Jun Xiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao, Jun","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129856253","display_name":"Yongliang Shen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shen, Yongliang","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5130059365"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.6945000290870667,"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":0.6945000290870667,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.02850000001490116,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.02319999970495701,"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/bootstrapping","display_name":"Bootstrapping (finance)","score":0.554099977016449},{"id":"https://openalex.org/keywords/generator","display_name":"Generator (circuit theory)","score":0.4814999997615814},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.4562999904155731},{"id":"https://openalex.org/keywords/term","display_name":"Term (time)","score":0.4050000011920929},{"id":"https://openalex.org/keywords/trap","display_name":"Trap (plumbing)","score":0.39419999718666077},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.38179999589920044},{"id":"https://openalex.org/keywords/majority-rule","display_name":"Majority rule","score":0.36719998717308044}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6848000288009644},{"id":"https://openalex.org/C207609745","wikidata":"https://www.wikidata.org/wiki/Q4944086","display_name":"Bootstrapping (finance)","level":2,"score":0.554099977016449},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.492900013923645},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.4814999997615814},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4562999904155731},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.43230000138282776},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.4050000011920929},{"id":"https://openalex.org/C121099081","wikidata":"https://www.wikidata.org/wiki/Q665580","display_name":"Trap (plumbing)","level":2,"score":0.39419999718666077},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.38179999589920044},{"id":"https://openalex.org/C153668964","wikidata":"https://www.wikidata.org/wiki/Q27636","display_name":"Majority rule","level":2,"score":0.36719998717308044},{"id":"https://openalex.org/C520049643","wikidata":"https://www.wikidata.org/wiki/Q189760","display_name":"Voting","level":3,"score":0.36250001192092896},{"id":"https://openalex.org/C2781316041","wikidata":"https://www.wikidata.org/wiki/Q1230584","display_name":"Diversity (politics)","level":2,"score":0.32429999113082886},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.31470000743865967},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.2971000075340271},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.27000001072883606},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.26350000500679016},{"id":"https://openalex.org/C100253034","wikidata":"https://www.wikidata.org/wiki/Q196372","display_name":"Systematic error","level":2,"score":0.25589999556541443},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.2515999972820282},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.2506999969482422}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.17775","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17775","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.17775","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17775","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Label-free":[0],"reinforcement":[1],"learning":[2],"enables":[3],"large":[4],"language":[5],"models":[6],"to":[7,38,136],"improve":[8],"reasoning":[9,127],"capabilities":[10,142],"without":[11],"ground-truth":[12],"supervision,":[13],"typically":[14],"by":[15,123],"treating":[16],"majority-voted":[17],"answers":[18],"as":[19,28],"pseudo-labels.":[20,96],"However,":[21],"we":[22,55],"identify":[23],"a":[24,58,61,100],"critical":[25],"failure":[26],"mode:":[27],"training":[29,84],"maximizes":[30],"self-consistency,":[31],"output":[32],"diversity":[33],"collapses,":[34],"causing":[35],"the":[36,49,74,85,88],"model":[37,63,115],"confidently":[39],"reinforce":[40],"systematic":[41],"errors":[42,94],"that":[43,103,118,140],"evade":[44],"detection.":[45],"We":[46],"term":[47],"this":[48],"consensus":[50],"trap.":[51],"To":[52],"escape":[53],"it,":[54],"propose":[56],"CoVerRL,":[57],"framework":[59],"where":[60],"single":[62],"alternates":[64],"between":[65],"generator":[66],"and":[67,113],"verifier":[68,90],"roles,":[69],"with":[70],"each":[71],"capability":[72],"bootstrapping":[73],"other.":[75],"Majority":[76],"voting":[77],"provides":[78],"noisy":[79],"but":[80],"informative":[81],"supervision":[82],"for":[83],"verifier,":[86],"while":[87],"improving":[89],"progressively":[91],"filters":[92],"self-consistent":[93],"from":[95,133],"This":[97],"co-evolution":[98],"creates":[99],"virtuous":[101],"cycle":[102],"maintains":[104],"high":[105],"reward":[106],"accuracy":[107,131],"throughout":[108],"training.":[109],"Experiments":[110],"across":[111],"Qwen":[112],"Llama":[114],"families":[116],"demonstrate":[117],"CoVerRL":[119],"outperforms":[120],"label-free":[121],"baselines":[122],"4.7-5.9%":[124],"on":[125],"mathematical":[126],"benchmarks.":[128],"Moreover,":[129],"self-verification":[130],"improves":[132],"around":[134],"55%":[135],"over":[137],"85%,":[138],"confirming":[139],"both":[141],"genuinely":[143],"co-evolve.":[144]},"counts_by_year":[],"updated_date":"2026-04-17T18:11:37.981687","created_date":"2026-03-20T00:00:00"}