{"id":"https://openalex.org/W4415245727","doi":"https://doi.org/10.48550/arxiv.2505.02865","title":"Accelerating Large Language Model Reasoning via Speculative Search","display_name":"Accelerating Large Language Model Reasoning via Speculative Search","publication_year":2025,"publication_date":"2025-05-03","ids":{"openalex":"https://openalex.org/W4415245727","doi":"https://doi.org/10.48550/arxiv.2505.02865"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2505.02865","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.02865","pdf_url":"https://arxiv.org/pdf/2505.02865","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2505.02865","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100698059","display_name":"Zhihai Wang","orcid":"https://orcid.org/0000-0002-9971-9167"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Wang, Zhihai","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100440145","display_name":"Jie Wang","orcid":"https://orcid.org/0000-0002-9663-3165"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jie","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120391877","display_name":"Jilai Pan","orcid":"https://orcid.org/0000-0002-3128-9025"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan, Jilai","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Xia, Xilin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xia, Xilin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031170152","display_name":"Hui\u2010Ling Zhen","orcid":"https://orcid.org/0000-0003-0310-3825"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhen, Huiling","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078949174","display_name":"Mingxuan Yuan","orcid":"https://orcid.org/0000-0002-2236-8784"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan, Mingxuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072138293","display_name":"Jianye Hao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao, Jianye","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5029102289","display_name":"Feng Wu","orcid":"https://orcid.org/0000-0003-3989-0509"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Feng","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5100698059"],"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9904000163078308,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9904000163078308,"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":0.9847999811172485,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.9745000004768372,"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":0.7353000044822693},{"id":"https://openalex.org/keywords/reasoning-system","display_name":"Reasoning system","score":0.6322000026702881},{"id":"https://openalex.org/keywords/non-monotonic-logic","display_name":"Non-monotonic logic","score":0.631600022315979},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6304000020027161},{"id":"https://openalex.org/keywords/limiting","display_name":"Limiting","score":0.589900016784668},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5570999979972839},{"id":"https://openalex.org/keywords/automated-reasoning","display_name":"Automated reasoning","score":0.5422999858856201},{"id":"https://openalex.org/keywords/model-based-reasoning","display_name":"Model-based reasoning","score":0.5307999849319458},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5188999772071838},{"id":"https://openalex.org/keywords/psychology-of-reasoning","display_name":"Psychology of reasoning","score":0.5166000127792358}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7573000192642212},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.7353000044822693},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.6322000026702881},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.631600022315979},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6304000020027161},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.589900016784668},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5570999979972839},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.5422999858856201},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.5307999849319458},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5188999772071838},{"id":"https://openalex.org/C183521366","wikidata":"https://www.wikidata.org/wiki/Q7256422","display_name":"Psychology of reasoning","level":4,"score":0.5166000127792358},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5026000142097473},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.4503999948501587},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.4472000002861023},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.43209999799728394},{"id":"https://openalex.org/C103057564","wikidata":"https://www.wikidata.org/wiki/Q4751139","display_name":"Analytic reasoning","level":3,"score":0.424699991941452},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.4097999930381775},{"id":"https://openalex.org/C97364631","wikidata":"https://www.wikidata.org/wiki/Q484284","display_name":"Deductive reasoning","level":2,"score":0.4066999852657318},{"id":"https://openalex.org/C166088908","wikidata":"https://www.wikidata.org/wiki/Q308495","display_name":"Abductive reasoning","level":2,"score":0.4034999907016754},{"id":"https://openalex.org/C115086926","wikidata":"https://www.wikidata.org/wiki/Q17004651","display_name":"Causal reasoning","level":3,"score":0.38089999556541443},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.3734999895095825},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.36419999599456787},{"id":"https://openalex.org/C36964233","wikidata":"https://www.wikidata.org/wiki/Q7920942","display_name":"Verbal reasoning","level":3,"score":0.3517000079154968},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.30480000376701355},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.30379998683929443},{"id":"https://openalex.org/C107848011","wikidata":"https://www.wikidata.org/wiki/Q4680756","display_name":"Adaptive reasoning","level":4,"score":0.28450000286102295},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.28290000557899475},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.27880001068115234},{"id":"https://openalex.org/C140843580","wikidata":"https://www.wikidata.org/wiki/Q840067","display_name":"Defeasible reasoning","level":2,"score":0.2768999934196472},{"id":"https://openalex.org/C43971567","wikidata":"https://www.wikidata.org/wiki/Q3142865","display_name":"Logical reasoning","level":2,"score":0.26190000772476196},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.25270000100135803}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2505.02865","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.02865","pdf_url":"https://arxiv.org/pdf/2505.02865","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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.2505.02865","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2505.02865","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2505.02865","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.02865","pdf_url":"https://arxiv.org/pdf/2505.02865","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Tree-search-based":[0],"reasoning":[1,7,21,39,61,88,123,151],"methods":[2,27],"have":[3,35],"significantly":[4,58,140],"enhanced":[5],"the":[6,16,112,126,132],"capability":[8],"of":[9,18,93,111],"large":[10,77,113,127],"language":[11],"models":[12,136],"(LLMs)":[13],"by":[14,62],"facilitating":[15],"exploration":[17],"multiple":[19],"intermediate":[20],"steps,":[22],"i.e.,":[23],"thoughts.":[24,89],"However,":[25],"these":[26],"suffer":[28],"from":[29],"substantial":[30],"inference":[31],"latency,":[32],"as":[33],"they":[34],"to":[36,72,125,146],"generate":[37],"numerous":[38],"thoughts,":[40],"severely":[41],"limiting":[42],"LLM":[43,60],"applicability.":[44],"To":[45],"address":[46],"this":[47],"challenge,":[48],"we":[49,117],"propose":[50],"a":[51,69,76,96],"novel":[52,97],"Speculative":[53],"Search":[54],"(SpecSearch)":[55],"framework":[56],"that":[57,110,119,138],"accelerates":[59],"optimizing":[63],"thought":[64,81],"generation.":[65],"Specifically,":[66],"SpecSearch":[67,94,120,139],"utilizes":[68],"small":[70],"model":[71,78],"strategically":[73],"collaborate":[74],"with":[75,149],"at":[79],"both":[80,131],"and":[82,134],"token":[83],"levels,":[84],"efficiently":[85],"generating":[86],"high-quality":[87],"The":[90],"major":[91],"pillar":[92],"is":[95],"quality-preserving":[98],"rejection":[99],"mechanism,":[100],"which":[101],"effectively":[102],"filters":[103],"out":[104],"thoughts":[105],"whose":[106],"quality":[107,124],"falls":[108],"below":[109],"model's":[114],"outputs.":[115],"Moreover,":[116],"show":[118],"preserves":[121],"comparable":[122,150],"model.":[128],"Experiments":[129],"on":[130],"Qwen":[133],"Llama":[135],"demonstrate":[137],"outperforms":[141],"state-of-the-art":[142],"approaches,":[143],"achieving":[144],"up":[145],"2.12$\\times$":[147],"speedup":[148],"quality.":[152]},"counts_by_year":[],"updated_date":"2026-04-16T08:26:57.006410","created_date":"2025-10-16T00:00:00"}