{"id":"https://openalex.org/W7128616912","doi":"https://doi.org/10.48550/arxiv.2602.10004","title":"ESTAR: Early-Stopping Token-Aware Reasoning For Efficient Inference","display_name":"ESTAR: Early-Stopping Token-Aware Reasoning For Efficient Inference","publication_year":2026,"publication_date":"2026-02-10","ids":{"openalex":"https://openalex.org/W7128616912","doi":"https://doi.org/10.48550/arxiv.2602.10004"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.10004","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","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":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5125670498","display_name":"Junda Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Wang, Junda","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125648174","display_name":"Zhichao Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Zhichao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125665030","display_name":"Dongxu Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Dongxu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042529162","display_name":"Sanjit Singh Batra","orcid":"https://orcid.org/0000-0001-9849-865X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Batra, Sanjit Singh","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5125641678","display_name":"Robert E. Tillman","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tillman, Robert E.","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5125670498"],"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.49149999022483826,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.49149999022483826,"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/T10028","display_name":"Topic Modeling","score":0.10010000318288803,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.08990000188350677,"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/opportunistic-reasoning","display_name":"Opportunistic reasoning","score":0.6635000109672546},{"id":"https://openalex.org/keywords/model-based-reasoning","display_name":"Model-based reasoning","score":0.6460000276565552},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6097999811172485},{"id":"https://openalex.org/keywords/reasoning-system","display_name":"Reasoning system","score":0.6033999919891357},{"id":"https://openalex.org/keywords/non-monotonic-logic","display_name":"Non-monotonic logic","score":0.5311999917030334},{"id":"https://openalex.org/keywords/case-based-reasoning","display_name":"Case-based reasoning","score":0.5149999856948853},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.4945000112056732},{"id":"https://openalex.org/keywords/classifier","display_name":"Classifier (UML)","score":0.4677000045776367}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7340999841690063},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.6635000109672546},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.6460000276565552},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6347000002861023},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6097999811172485},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.6033999919891357},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.5311999917030334},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.5149999856948853},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.4945000112056732},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.4677000045776367},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.45399999618530273},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4221999943256378},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.41920000314712524},{"id":"https://openalex.org/C97364631","wikidata":"https://www.wikidata.org/wiki/Q484284","display_name":"Deductive reasoning","level":2,"score":0.35600000619888306},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.3424000144004822},{"id":"https://openalex.org/C103057564","wikidata":"https://www.wikidata.org/wiki/Q4751139","display_name":"Analytic reasoning","level":3,"score":0.34139999747276306},{"id":"https://openalex.org/C166088908","wikidata":"https://www.wikidata.org/wiki/Q308495","display_name":"Abductive reasoning","level":2,"score":0.3379000127315521},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.29919999837875366},{"id":"https://openalex.org/C115086926","wikidata":"https://www.wikidata.org/wiki/Q17004651","display_name":"Causal reasoning","level":3,"score":0.2879999876022339},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.2842999994754791},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.26489999890327454},{"id":"https://openalex.org/C36964233","wikidata":"https://www.wikidata.org/wiki/Q7920942","display_name":"Verbal reasoning","level":3,"score":0.25850000977516174}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.10004","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.10004","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.10004","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":"pmh:doi:10.48550/arxiv.2602.10004","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"reasoning":[1,17,38,56,89,95,127],"models":[2],"(LRMs)":[3],"achieve":[4],"state-of-the-art":[5],"performance":[6],"by":[7,97],"generating":[8],"long":[9],"chains-of-thought,":[10],"but":[11],"often":[12],"waste":[13],"computation":[14],"on":[15,87],"redundant":[16],"after":[18],"the":[19],"correct":[20],"answer":[21],"has":[22],"already":[23],"been":[24],"reached.":[25],"We":[26],"introduce":[27],"Early-Stopping":[28],"for":[29,125],"Token-Aware":[30],"Reasoning":[31],"(ESTAR),":[32],"which":[33],"detects":[34],"and":[35,71],"reduces":[36,94],"such":[37],"redundancy":[39],"to":[40,64,67,102],"improve":[41],"efficiency":[42,128],"without":[43],"sacrificing":[44],"accuracy.":[45],"Our":[46],"method":[47],"combines":[48],"(i)":[49],"a":[50,120],"trajectory-based":[51],"classifier":[52],"that":[53,76,92],"identifies":[54],"when":[55],"can":[57],"be":[58],"safely":[59],"stopped,":[60],"(ii)":[61],"supervised":[62],"fine-tuning":[63],"teach":[65],"LRMs":[66],"propose":[68],"self-generated":[69,80],"signals,":[70],"(iii)":[72],"-aware":[73],"reinforcement":[74],"learning":[75],"truncates":[77],"rollouts":[78],"at":[79],"stop":[81],"points":[82],"with":[83,110],"compute-aware":[84],"rewards.":[85],"Experiments":[86],"four":[88],"datasets":[90],"show":[91],"ESTAR":[93],"length":[96],"about":[98],"3.7x":[99],"(from":[100],"4,799":[101],"1,290)":[103],"while":[104],"preserving":[105],"accuracy":[106],"(74.9%":[107],"vs.":[108],"74.2%),":[109],"strong":[111],"cross-domain":[112],"generalization.":[113],"These":[114],"results":[115],"highlight":[116],"early":[117],"stopping":[118],"as":[119],"simple":[121],"yet":[122],"powerful":[123],"mechanism":[124],"improving":[126],"in":[129],"LRMs.":[130]},"counts_by_year":[],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2026-02-12T00:00:00"}