{"id":"https://openalex.org/W7163139123","doi":"https://doi.org/10.48550/arxiv.2606.00487","title":"TAPS: Target-Aware Prefix Tree Selection for Diffusion-Drafted Speculative Decoding","display_name":"TAPS: Target-Aware Prefix Tree Selection for Diffusion-Drafted Speculative Decoding","publication_year":2026,"publication_date":"2026-05-30","ids":{"openalex":"https://openalex.org/W7163139123","doi":"https://doi.org/10.48550/arxiv.2606.00487"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.00487","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.00487","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":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.2606.00487","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137656393","display_name":"Zhuoyu Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zhuoyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137647141","display_name":"Junnan Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Junnan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137689851","display_name":"Xinyu Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Xinyu","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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.30149999260902405,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.30149999260902405,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.15440000593662262,"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":0.1151999980211258,"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/prefix","display_name":"Prefix","score":0.6434000134468079},{"id":"https://openalex.org/keywords/lossless-compression","display_name":"Lossless compression","score":0.6236000061035156},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6205999851226807},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.6047000288963318},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5091999769210815},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.5080000162124634},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.5031999945640564},{"id":"https://openalex.org/keywords/tree","display_name":"Tree (set theory)","score":0.45320001244544983}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7827000021934509},{"id":"https://openalex.org/C141603448","wikidata":"https://www.wikidata.org/wiki/Q134830","display_name":"Prefix","level":2,"score":0.6434000134468079},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.6236000061035156},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6205999851226807},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.6047000288963318},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5091999769210815},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5080000162124634},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.5031999945640564},{"id":"https://openalex.org/C113174947","wikidata":"https://www.wikidata.org/wiki/Q2859736","display_name":"Tree (set theory)","level":2,"score":0.45320001244544983},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4260999858379364},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.41260001063346863},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3968999981880188},{"id":"https://openalex.org/C190290938","wikidata":"https://www.wikidata.org/wiki/Q387015","display_name":"Trie","level":3,"score":0.37630000710487366},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.33980000019073486},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.3361000120639801},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.3327000141143799},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.30880001187324524},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3037000000476837},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3001999855041504},{"id":"https://openalex.org/C163797641","wikidata":"https://www.wikidata.org/wiki/Q2067937","display_name":"Tree structure","level":3,"score":0.2874000072479248},{"id":"https://openalex.org/C147297375","wikidata":"https://www.wikidata.org/wiki/Q6674930","display_name":"Look-ahead","level":2,"score":0.26759999990463257},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.25440001487731934},{"id":"https://openalex.org/C3017489831","wikidata":"https://www.wikidata.org/wiki/Q2393193","display_name":"Running time","level":2,"score":0.25440001487731934}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.00487","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.00487","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.00487","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.00487","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":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/17","display_name":"Partnerships for the goals","score":0.40858739614486694}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Using":[0],"a":[1,8,22,40,57,79,101,117,122],"diffusion":[2,26,108],"model":[3,142],"for":[4,11],"parallel":[5],"drafting":[6,30],"is":[7,76,170],"promising":[9],"approach":[10],"speculative":[12],"decoding.":[13],"By":[14],"predicting":[15],"tokens":[16],"at":[17,172],"multiple":[18],"future":[19],"positions":[20],"in":[21,60],"single":[23,41],"forward":[24],"pass,":[25],"drafters":[27],"substantially":[28],"reduce":[29],"latency.":[31,54],"However,":[32],"this":[33],"shifts":[34],"the":[35,70,127,134],"bottleneck":[36],"to":[37,149],"verification:":[38],"verifying":[39,47],"sequence":[42],"limits":[43],"acceptance":[44,93,112],"length,":[45],"while":[46],"large":[48],"draft":[49,135],"trees":[50],"incurs":[51],"excessive":[52],"target-model":[53],"We":[55],"identify":[56],"key":[58],"mismatch":[59],"existing":[61,64],"draft-tree":[62],"methods:":[63],"diffusion-tree":[65],"methods":[66],"rank":[67],"nodes":[68],"by":[69,163],"marginal":[71],"probability,":[72],"ignoring":[73],"that":[74,106,145],"verification":[75,124],"prefix-conditioned.":[77],"As":[78],"result,":[80],"they":[81],"may":[82],"verify":[83],"unreachable":[84],"descendants":[85],"of":[86],"rejected":[87],"prefixes,":[88],"increasing":[89],"latency":[90],"with":[91],"limited":[92],"gains.":[94],"To":[95],"address":[96],"this,":[97],"we":[98],"propose":[99],"TAPS,":[100],"target-aware":[102],"prefix":[103],"selection":[104],"method":[105],"turns":[107],"marginals":[109],"into":[110],"path-conditioned":[111],"estimates.":[113],"TAPS":[114,146],"then":[115],"selects":[116],"compact":[118],"prefix-closed":[119],"subtree":[120],"under":[121],"fixed":[123],"budget,":[125],"improving":[126],"acceptance-cost":[128],"tradeoff":[129],"rather":[130],"than":[131],"simply":[132],"expanding":[133],"tree.":[136],"Experiments":[137],"across":[138],"diverse":[139],"datasets":[140],"and":[141,161,165],"families":[143],"demonstrate":[144],"achieves":[147],"up":[148],"7.9x":[150],"lossless":[151],"end-to-end":[152],"speedup":[153],"over":[154],"vanilla":[155],"autoregressive":[156],"decoding,":[157],"outperforming":[158],"state-of-the-art":[159],"DFlash":[160],"DDTree":[162],"1.36x":[164],"1.74x":[166],"respectively.":[167],"Our":[168],"work":[169],"available":[171],"https://anonymous.4open.science/r/TAPS-EMNLP2026-53DD":[173]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-03T00:00:00"}