{"id":"https://openalex.org/W7134907704","doi":"https://doi.org/10.48550/arxiv.2603.08899","title":"ConFu: Contemplate the Future for Better Speculative Sampling","display_name":"ConFu: Contemplate the Future for Better Speculative Sampling","publication_year":2026,"publication_date":"2026-03-09","ids":{"openalex":"https://openalex.org/W7134907704","doi":"https://doi.org/10.48550/arxiv.2603.08899"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.08899","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.08899","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.2603.08899","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128724089","display_name":"Zongyue Qin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qin, Zongyue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114129913","display_name":"Raghavv Goel","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Goel, Raghavv","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018463719","display_name":"Mukul Gagrani","orcid":"https://orcid.org/0000-0003-0483-1090"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gagrani, Mukul","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128708241","display_name":"Risheek Garrepalli","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Garrepalli, Risheek","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128723541","display_name":"Mingu Lee","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lee, Mingu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128775296","display_name":"Yizhou Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Yizhou","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":0.27649998664855957,"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":0.27649998664855957,"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/T13910","display_name":"Computational and Text Analysis Methods","score":0.11980000138282776,"subfield":{"id":"https://openalex.org/subfields/3300","display_name":"General Social Sciences"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T13629","display_name":"Text Readability and Simplification","score":0.059700001031160355,"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/security-token","display_name":"Security token","score":0.7839000225067139},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.690500020980835},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.6510999798774719},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5236999988555908},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.40529999136924744},{"id":"https://openalex.org/keywords/downstream","display_name":"Downstream (manufacturing)","score":0.3853999972343445},{"id":"https://openalex.org/keywords/bridge","display_name":"Bridge (graph theory)","score":0.3230000138282776}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7839999794960022},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7839000225067139},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.690500020980835},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.6510999798774719},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5236999988555908},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4185999929904938},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.40529999136924744},{"id":"https://openalex.org/C2776207758","wikidata":"https://www.wikidata.org/wiki/Q5303302","display_name":"Downstream (manufacturing)","level":2,"score":0.3853999972343445},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35580000281333923},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.3230000138282776},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.29600000381469727},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.2912999987602234},{"id":"https://openalex.org/C73602740","wikidata":"https://www.wikidata.org/wiki/Q7795822","display_name":"Thompson sampling","level":3,"score":0.289900004863739},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.275299996137619},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.2752000093460083},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.2612999975681305},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.2612999975681305},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.2574000060558319}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.08899","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.08899","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.2603.08899","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.08899","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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Speculative":[0],"decoding":[1,95,198],"has":[2],"emerged":[3],"as":[4,50],"a":[5,92,133,147,204],"powerful":[6],"approach":[7],"to":[8,20,75,101,121,140,195],"accelerate":[9],"large":[10],"language":[11],"model":[12,80,120,128],"(LLM)":[13],"inference":[14],"by":[15,28,62,173,179],"employing":[16],"lightweight":[17],"draft":[18,44,58,99,119],"models":[19,59,100],"propose":[21,87],"candidate":[22],"tokens":[23,112],"that":[24,97,116,158],"are":[25],"subsequently":[26],"verified":[27],"the":[29,40,43,51,69,78,90,103,118,126,193],"target":[30,79,127],"model.":[31,45],"The":[32],"effectiveness":[33],"of":[34,42,106],"this":[35,84],"paradigm":[36],"critically":[37],"depends":[38],"on":[39,68,175,182],"quality":[41],"While":[46],"recent":[47],"advances":[48],"such":[49],"EAGLE":[52],"series":[53],"achieve":[54],"state-of-the-art":[55],"speedup,":[56],"existing":[57],"remain":[60],"limited":[61],"error":[63],"accumulation:":[64],"they":[65],"condition":[66],"only":[67],"current":[70],"prefix,":[71],"causing":[72],"their":[73],"predictions":[74],"drift":[76],"from":[77,125],"over":[81,171],"steps.":[82],"In":[83],"work,":[85],"we":[86],"\\textbf{ConFu}":[88],"(Contemplate":[89],"Future),":[91],"novel":[93],"speculative":[94,197],"framework":[96,149],"enables":[98],"anticipate":[102],"future":[104,143,155,161],"direction":[105,206],"generation.":[107],"ConFu":[108,163],"introduces":[109],"(i)":[110],"contemplate":[111,135],"and":[113,145,154,168,178],"soft":[114],"prompts":[115],"allow":[117],"leverage":[122],"future-oriented":[123],"signals":[124],"at":[129],"negligible":[130],"cost,":[131],"(ii)":[132],"dynamic":[134],"token":[136,152,165],"mechanism":[137],"with":[138,150,199],"MoE":[139],"enable":[141],"context-aware":[142],"prediction,":[144],"(iii)":[146],"training":[148],"anchor":[151],"sampling":[153],"prediction":[156],"replication":[157],"learns":[159],"robust":[160],"prediction.":[162],"improves":[164],"acceptance":[166],"rates":[167],"generation":[169],"speed":[170],"EAGLE-3":[172],"8--11\\%":[174],"Llama-3":[176],"3B/8B":[177],"approximately":[180],"20\\%":[181],"Qwen-3":[183],"4B":[184],"across":[185],"downstream":[186],"tasks.":[187],"We":[188],"believe":[189],"our":[190],"work":[191],"is":[192],"first":[194],"bridge":[196],"continuous":[200],"reasoning":[201],"tokens,":[202],"offering":[203],"new":[205],"for":[207],"accelerating":[208],"LLM":[209],"inference.":[210]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-12T00:00:00"}