{"id":"https://openalex.org/W4416956409","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1650","title":"Beyond High-Entropy Exploration: Correctness-Aware Low-Entropy Segment-Based Advantage Shaping for Reasoning LLMs","display_name":"Beyond High-Entropy Exploration: Correctness-Aware Low-Entropy Segment-Based Advantage Shaping for Reasoning LLMs","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4416956409","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1650"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1650","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1650","pdf_url":"https://aclanthology.org/2026.findings-acl.1650.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":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.1650.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5109533436","display_name":"X.F. Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinzhu Chen","raw_affiliation_strings":["Beijing University of Posts and Telecommunications , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications , China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100672848","display_name":"Xinghua Li","orcid":"https://orcid.org/0000-0002-3275-2755"},"institutions":[{"id":"https://openalex.org/I146563203","display_name":"University of International Business and Economics","ror":"https://ror.org/05khqpb71","country_code":"CN","type":"education","lineage":["https://openalex.org/I146563203"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xuesheng Li","raw_affiliation_strings":["School of Artificial Intelligence and Data Science , University of International Business and Economics , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Artificial Intelligence and Data Science , University of International Business and Economics , China","institution_ids":["https://openalex.org/I146563203"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Zhongxiang Sun","orcid":null},"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":"Zhongxiang Sun","raw_affiliation_strings":["Gaoling School of Artificial Intelligence , Renmin University of China , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence , Renmin University of China , China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5008201587","display_name":"Wen Yu","orcid":"https://orcid.org/0000-0002-9540-7924"},"institutions":[{"id":"https://openalex.org/I146563203","display_name":"University of International Business and Economics","ror":"https://ror.org/05khqpb71","country_code":"CN","type":"education","lineage":["https://openalex.org/I146563203"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weijie Yu","raw_affiliation_strings":["School of Artificial Intelligence and Data Science , University of International Business and Economics , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Artificial Intelligence and Data Science , University of International Business and Economics , China","institution_ids":["https://openalex.org/I146563203"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"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":0.03117512,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"32970","last_page":"32984"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.5350000262260437,"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":0.5350000262260437,"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":0.10000000149011612,"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":0.0812000036239624,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6420999765396118},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5777999758720398},{"id":"https://openalex.org/keywords/encode","display_name":"ENCODE","score":0.5015000104904175},{"id":"https://openalex.org/keywords/verifiable-secret-sharing","display_name":"Verifiable secret sharing","score":0.4772000014781952},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.4458000063896179},{"id":"https://openalex.org/keywords/through-the-lens-metering","display_name":"Through-the-lens metering","score":0.3698999881744385}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6812999844551086},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6420999765396118},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5777999758720398},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.5015000104904175},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4932999908924103},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.4772000014781952},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.4458000063896179},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.3840999901294708},{"id":"https://openalex.org/C43091099","wikidata":"https://www.wikidata.org/wiki/Q1067788","display_name":"Through-the-lens metering","level":3,"score":0.3698999881744385},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.36090001463890076},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.34369999170303345},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3424000144004822},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3163999915122986},{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.31529998779296875},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.31209999322891235},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.27799999713897705},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.26429998874664307}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1650","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1650","pdf_url":"https://aclanthology.org/2026.findings-acl.1650.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"},{"id":"pmh:oai:arXiv.org:2512.00908","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.00908","pdf_url":"https://arxiv.org/pdf/2512.00908","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,"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.2512.00908","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.00908","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,"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":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.1650","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1650","pdf_url":"https://aclanthology.org/2026.findings-acl.1650.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":[{"id":"https://openalex.org/G2412026816","display_name":null,"funder_award_id":"24QN06","funder_id":"https://openalex.org/F4320335787","funder_display_name":"Fundamental Research Funds for the Central Universities"},{"id":"https://openalex.org/G7601973381","display_name":null,"funder_award_id":"24PYTS22","funder_id":"https://openalex.org/F4320335787","funder_display_name":"Fundamental Research Funds for the Central Universities"}],"funders":[{"id":"https://openalex.org/F4320312338","display_name":"University of International Business and Economics","ror":"https://ror.org/05khqpb71"},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320335787","display_name":"Fundamental Research Funds for the Central Universities","ror":null}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416956409.pdf","grobid_xml":"https://content.openalex.org/works/W4416956409.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"Learning":[1],"with":[2,75],"Verifiable":[3],"Rewards":[4],"(RLVR)":[5],"has":[6],"become":[7],"a":[8,44,94],"central":[9],"approach":[10],"for":[11],"improving":[12],"the":[13,39,65,129,161],"reasoning":[14,45],"ability":[15],"of":[16,43,48,67,135,160],"large":[17],"language":[18],"models.Recent":[19],"work":[20],"studies":[21],"RLVR":[22],"through":[23],"token":[24],"entropy,":[25],"arguing":[26],"that":[27,41,51,64,98],"highentropy":[28,126],"tokens":[29],"drive":[30],"exploration":[31,127],"and":[32,54,59,118,137,151],"should":[33],"receive":[34],"stronger":[35,158],"updates.However,":[36],"they":[37],"overlook":[38],"fact":[40],"most":[42],"trajectory":[46],"consists":[47],"low-entropy":[49,68],"segments":[50,69,107,120],"encode":[52],"stable":[53,84],"reusable":[55],"structural":[56],"patterns.Through":[57],"qualitative":[58],"quantitative":[60],"analyses,":[61],"we":[62,91],"find":[63],"overlap":[66],"across":[70,148],"correct":[71,110],"responses":[72,82],"strongly":[73],"correlates":[74],"model":[76],"accuracy,":[77],"while":[78,124],"overlaps":[79],"involving":[80],"incorrect":[81,116],"exhibit":[83],"but":[85,155],"unproductive":[86],"patterns.Motivated":[87],"by":[88,122],"these":[89],"findings,":[90],"propose":[92],"LESS,":[93],"correctnessaware":[95],"reinforcement":[96],"framework":[97],"performs":[99],"fine-grained":[100],"advantage":[101],"modulation":[102],"over":[103,144],"lowentropy":[104],"segments.LESS":[105],"amplifies":[106],"unique":[108,114],"to":[109,115],"responses,":[111],"suppresses":[112],"those":[113],"ones,":[117],"neutralizes":[119],"shared":[121],"both,":[123],"preserving":[125],"in":[128],"underlying":[130],"RL":[131,146],"algorithm.Instantiated":[132],"on":[133],"top":[134],"GRPO":[136],"GSPO,":[138],"LESS":[139],"not":[140],"only":[141],"improves":[142],"accuracy":[143],"strong":[145],"baselines":[147],"three":[149],"backbones":[150],"six":[152],"math":[153],"benchmarks,":[154],"also":[156],"achieves":[157],"robustness":[159],"performance":[162],"floor.":[163]},"counts_by_year":[],"updated_date":"2026-08-18T07:49:30.821534","created_date":"2025-12-03T00:00:00"}