{"id":"https://openalex.org/W4416242513","doi":"https://doi.org/10.48550/arxiv.2508.07534","title":"From Trial-and-Error to Improvement: A Systematic Analysis of LLM Exploration Mechanisms in RLVR","display_name":"From Trial-and-Error to Improvement: A Systematic Analysis of LLM Exploration Mechanisms in RLVR","publication_year":2025,"publication_date":"2025-08-11","ids":{"openalex":"https://openalex.org/W4416242513","doi":"https://doi.org/10.48550/arxiv.2508.07534"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2508.07534","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.07534","pdf_url":"https://arxiv.org/pdf/2508.07534","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2508.07534","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5006730106","display_name":"Jia Wei Deng","orcid":"https://orcid.org/0000-0002-9251-6818","observed_orcids":["https://orcid.org/0000-0002-9251-6818"]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Deng, Jia","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Chen, Jie","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen, Jie","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Chen, Zhipeng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen, Zhipeng","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126801103","display_name":"Daixuan Cheng","orcid":"https://orcid.org/0009-0002-3548-3391","observed_orcids":["https://orcid.org/0009-0002-3548-3391"]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cheng, Daixuan","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120402710","display_name":"Fei Bai","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bai, Fei","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032904822","display_name":"B. X. Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhang, Beichen","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120450055","display_name":"Yinqian Min","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Min, Yinqian","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Gao, Yanzipeng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Gao, Yanzipeng","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037145565","display_name":"Wayne Xin Zhao","orcid":"https://orcid.org/0000-0002-8333-6196","observed_orcids":["https://orcid.org/0000-0002-8333-6196"]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhao, Wayne Xin","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"last","author":{"id":null,"display_name":"Wen, Ji-Rong","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wen, Ji-Rong","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China","institution_ids":["https://openalex.org/I78988378"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I78988378"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":21,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.7226999998092651,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.7226999998092651,"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":0.10379999876022339,"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":0.05510000139474869,"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-with-verifiable-rewards","display_name":"reinforcement learning with verifiable rewards","score":0.9860000014305115},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.9150000214576721},{"id":"https://openalex.org/keywords/exploration-strategy","display_name":"exploration strategy","score":0.6759999990463257},{"id":"https://openalex.org/keywords/reasoning-chains","display_name":"reasoning chains","score":0.8420000076293945}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6690000295639038},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5871999859809875},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5641000270843506},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.4602999985218048},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.40230000019073486},{"id":"https://openalex.org/C539667460","wikidata":"https://www.wikidata.org/wiki/Q2414942","display_name":"Management science","level":1,"score":0.39579999446868896},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.38510000705718994},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.38119998574256897},{"id":"https://openalex.org/C166052673","wikidata":"https://www.wikidata.org/wiki/Q83021","display_name":"Empirical evidence","level":2,"score":0.3653999865055084},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.3379000127315521},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.3287000060081482},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.32679998874664307},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.32580000162124634},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3018999993801117},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.29109999537467957},{"id":"https://openalex.org/C104060986","wikidata":"https://www.wikidata.org/wiki/Q180046","display_name":"Space exploration","level":2,"score":0.2842000126838684},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.2824000120162964},{"id":"https://openalex.org/C174998907","wikidata":"https://www.wikidata.org/wiki/Q357662","display_name":"Work in process","level":2,"score":0.2736000120639801},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2606000006198883},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.2583000063896179},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.25519999861717224}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2508.07534","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.07534","pdf_url":"https://arxiv.org/pdf/2508.07534","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2508.07534","doi":"https://doi.org/10.48550/arxiv.2508.07534","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2508.07534","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2508.07534","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.07534","pdf_url":"https://arxiv.org/pdf/2508.07534","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416242513.pdf","grobid_xml":"https://content.openalex.org/works/W4416242513.grobid-xml"},"referenced_works_count":16,"referenced_works":["https://openalex.org/W2736601468","https://openalex.org/W3000779003","https://openalex.org/W4404987838","https://openalex.org/W4406961529","https://openalex.org/W4414859475","https://openalex.org/W4414896318","https://openalex.org/W4414898977","https://openalex.org/W4416048648","https://openalex.org/W4416600559","https://openalex.org/W4417103130","https://openalex.org/W4417438207","https://openalex.org/W6891877852","https://openalex.org/W6948538263","https://openalex.org/W7137809993","https://openalex.org/W7138345725","https://openalex.org/W7196953391"],"related_works":["https://openalex.org/W7084164689","https://openalex.org/W4417531499","https://openalex.org/W7204597172","https://openalex.org/W4416105072","https://openalex.org/W4416373918","https://openalex.org/W7160960353","https://openalex.org/W4414979505","https://openalex.org/W7154586467","https://openalex.org/W4416156365","https://openalex.org/W4414970985"],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1],"with":[2,128],"verifiable":[3],"rewards":[4],"(RLVR)":[5],"has":[6,52],"emerged":[7],"as":[8],"a":[9,41,70,137],"powerful":[10],"paradigm":[11],"for":[12,140],"enhancing":[13],"the":[14,57],"reasoning":[15,38],"capabilities":[16],"of":[17,73],"large":[18],"language":[19],"models":[20],"(LLMs).":[21],"Unlike":[22],"traditional":[23],"RL":[24,110],"approaches,":[25],"RLVR":[26,142],"leverages":[27],"rule-based":[28],"feedback":[29],"to":[30,91,115,135],"guide":[31],"LLMs":[32],"in":[33,76],"generating":[34],"and":[35,105,108],"refining":[36],"complex":[37],"chains":[39],"--":[40],"process":[42],"critically":[43],"dependent":[44],"on":[45],"effective":[46],"exploration":[47,62,74,83,118],"strategies.":[48],"While":[49],"prior":[50],"work":[51,133],"demonstrated":[53],"RLVR's":[54],"empirical":[55,130],"success,":[56],"fundamental":[58],"mechanisms":[59],"governing":[60],"LLMs'":[61,93],"behaviors":[63],"remain":[64],"underexplored.":[65],"This":[66],"technical":[67],"report":[68],"presents":[69],"systematic":[71],"investigation":[72],"capacities":[75],"RLVR,":[77],"covering":[78],"four":[79],"main":[80],"aspects:":[81],"(1)":[82],"space":[84],"shaping,":[85],"where":[86],"we":[87],"develop":[88],"quantitative":[89],"metrics":[90],"characterize":[92],"capability":[94],"boundaries;":[95],"(2)":[96],"entropy-performance":[97],"exchange,":[98],"analyzed":[99],"across":[100],"training":[101],"stages,":[102],"individual":[103],"instances,":[104],"token-level":[106],"patterns;":[107],"(3)":[109],"performance":[111],"optimization,":[112],"examining":[113],"methods":[114],"effectively":[116],"translate":[117],"gains":[119],"into":[120],"measurable":[121],"improvements.":[122],"By":[123],"unifying":[124],"previously":[125],"identified":[126],"insights":[127],"new":[129],"evidence,":[131],"this":[132],"aims":[134],"provide":[136],"foundational":[138],"framework":[139],"advancing":[141],"systems.":[143]},"counts_by_year":[{"year":2026,"cited_by_count":15},{"year":2025,"cited_by_count":6}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-10-10T00:00:00"}