{"id":"https://openalex.org/W7160918851","doi":"https://doi.org/10.48550/arxiv.2605.09806","title":"LEAD: Length-Efficient Adaptive and Dynamic Reasoning for Large Language Models","display_name":"LEAD: Length-Efficient Adaptive and Dynamic Reasoning for Large Language Models","publication_year":2026,"publication_date":"2026-05-10","ids":{"openalex":"https://openalex.org/W7160918851","doi":"https://doi.org/10.48550/arxiv.2605.09806"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.09806","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.09806","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.09806","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135985664","display_name":"Songtao Wei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Songtao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135963773","display_name":"Yi Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135976066","display_name":"Zhikai Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Zhikai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135962667","display_name":"Xu Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066436066","display_name":"Yuede Ji","orcid":"https://orcid.org/0000-0002-2419-6592"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ji, Yuede","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025726299","display_name":"Guanpeng Li","orcid":"https://orcid.org/0000-0001-7773-7826"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Guanpeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135976958","display_name":"Feng Chen","orcid":"https://orcid.org/0000-0003-4432-6005"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Feng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135916865","display_name":"Carl Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Carl","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135969149","display_name":"Zhichun Guo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Zhichun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5048972267","display_name":"Bingzhe Li","orcid":"https://orcid.org/0000-0002-5815-9706"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Bingzhe","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":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":0.34929999709129333,"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":0.34929999709129333,"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":0.2754000127315521,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.039799999445676804,"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/correctness","display_name":"Correctness","score":0.6779999732971191},{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.6620000004768372},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6007999777793884},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5182999968528748},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.3716999888420105},{"id":"https://openalex.org/keywords/case-based-reasoning","display_name":"Case-based reasoning","score":0.36559998989105225},{"id":"https://openalex.org/keywords/base","display_name":"Base (topology)","score":0.3578999936580658},{"id":"https://openalex.org/keywords/automated-reasoning","display_name":"Automated reasoning","score":0.31779998540878296}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7376000285148621},{"id":"https://openalex.org/C55439883","wikidata":"https://www.wikidata.org/wiki/Q360812","display_name":"Correctness","level":2,"score":0.6779999732971191},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.6620000004768372},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6223999857902527},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6007999777793884},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5182999968528748},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.46889999508857727},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.3716999888420105},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.36559998989105225},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.3578999936580658},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.31779998540878296},{"id":"https://openalex.org/C46355384","wikidata":"https://www.wikidata.org/wiki/Q726686","display_name":"Compromise","level":2,"score":0.31189998984336853},{"id":"https://openalex.org/C196340769","wikidata":"https://www.wikidata.org/wiki/Q7698910","display_name":"Temporal difference learning","level":3,"score":0.27559998631477356},{"id":"https://openalex.org/C125014702","wikidata":"https://www.wikidata.org/wiki/Q4680749","display_name":"Adaptive learning","level":2,"score":0.2750999927520752},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.27070000767707825},{"id":"https://openalex.org/C168031717","wikidata":"https://www.wikidata.org/wiki/Q1530280","display_name":"Balance (ability)","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.26829999685287476},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.26809999346733093},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2614000141620636},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.2599000036716461},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.2590000033378601}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.09806","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.09806","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.09806","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.09806","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":"Preprint"},"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":{"Large":[0],"reasoning":[1,16,69,169],"models,":[2],"such":[3],"as":[4,14],"OpenAI":[5],"o1":[6],"and":[7,34,61,67,80,91,103,163,176],"DeepSeek-R1,":[8],"tend":[9],"to":[10,132],"become":[11],"increasingly":[12],"verbose":[13],"their":[15],"capabilities":[17],"improve.":[18],"These":[19],"inflated":[20],"Chain-of-Thought":[21],"(CoT)":[22],"trajectories":[23],"often":[24],"exceed":[25],"what":[26],"the":[27,56,119,133,149,173,189],"underlying":[28],"problems":[29],"require,":[30],"wasting":[31],"compute,":[32],"latency,":[33],"context":[35],"budgets.":[36],"While":[37],"introducing":[38],"length-based":[39],"efficiency":[40,62,157],"rewards":[41],"during":[42],"reinforcement":[43],"learning":[44,136],"offers":[45],"a":[46,86,106,126,155],"natural":[47],"remedy,":[48],"existing":[49],"methods":[50,182],"struggle":[51],"with":[52,112],"two":[53],"fundamental":[54],"challenges:":[55],"optimal":[57],"balance":[58],"between":[59,88],"correctness":[60],"is":[63],"non-stationary":[64],"throughout":[65],"training,":[66],"intrinsic":[68],"budgets":[70],"vary":[71],"drastically":[72],"across":[73],"problems.":[74],"Relying":[75],"on":[76,148,166],"static":[77,110],"reward":[78,158],"weights":[79],"global":[81],"length":[82,145],"constraints":[83],"inevitably":[84],"forces":[85],"compromise":[87],"degraded":[89],"accuracy":[90,175],"unrealized":[92],"compression.":[93],"To":[94],"overcome":[95],"these":[96],"limitations,":[97],"we":[98],"propose":[99],"LEAD":[100,116,171],"(Length-Efficient":[101],"Adaptive":[102],"Dynamic":[104],"reasoning),":[105],"method":[107],"that":[108,159],"replaces":[109],"heuristics":[111],"online,":[113],"self-adaptive":[114],"mechanisms.":[115],"dynamically":[117],"calibrates":[118],"correctness-efficiency":[120],"trade-off":[121],"at":[122],"each":[123],"step":[124],"using":[125],"Potential-Scaled":[127],"Instability,":[128],"directing":[129],"optimization":[130],"capacity":[131],"most":[134],"informative":[135],"signal.":[137],"Furthermore,":[138],"it":[139],"estimates":[140],"an":[141],"adaptive":[142],"per-problem":[143],"target":[144],"online":[146],"based":[147],"model's":[150],"own":[151],"correct":[152],"rollouts,":[153],"applying":[154],"symmetric":[156],"penalizes":[160],"both":[161],"overthinking":[162],"over-compression.":[164],"Evaluated":[165],"five":[167],"mathematical":[168],"benchmarks,":[170],"achieves":[172],"highest":[174],"Accuracy-Efficiency":[177],"Score":[178],"among":[179],"RL-trained":[180],"efficient-reasoning":[181],"while":[183],"producing":[184],"substantially":[185],"shorter":[186],"outputs":[187],"than":[188],"base":[190],"model.":[191]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-13T00:00:00"}