{"id":"https://openalex.org/W7163407797","doi":"https://doi.org/10.48550/arxiv.2606.02955","title":"Fast-dLLM++: Fr\u00e9chet Profile Decoding for Faster Diffusion LLM Inference","display_name":"Fast-dLLM++: Fr\u00e9chet Profile Decoding for Faster Diffusion LLM Inference","publication_year":2026,"publication_date":"2026-06-01","ids":{"openalex":"https://openalex.org/W7163407797","doi":"https://doi.org/10.48550/arxiv.2606.02955"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.02955","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.02955","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.2606.02955","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5041090300","display_name":"Siva Rajesh Kasa","orcid":"https://orcid.org/0000-0003-1509-4313"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kasa, Siva Rajesh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137723054","display_name":"Yasong Dai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dai, Yasong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109167303","display_name":"Sumit Negi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Negi, Sumit","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5101819061","display_name":"Hongdong Li","orcid":"https://orcid.org/0000-0003-4125-1554"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Hongdong","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/T10181","display_name":"Natural Language Processing Techniques","score":0.3278000056743622,"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":0.3278000056743622,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.13009999692440033,"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":0.10700000077486038,"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/decoding-methods","display_name":"Decoding methods","score":0.710099995136261},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5889000296592712},{"id":"https://openalex.org/keywords/commit","display_name":"Commit","score":0.5608999729156494},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5579000115394592},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5532000064849854},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5263000130653381},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4702000021934509},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.46860000491142273},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.43869999051094055},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.4120999872684479}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.745199978351593},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.710099995136261},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5889000296592712},{"id":"https://openalex.org/C153180980","wikidata":"https://www.wikidata.org/wiki/Q19776675","display_name":"Commit","level":2,"score":0.5608999729156494},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5579000115394592},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5532000064849854},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5263000130653381},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4941999912261963},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4702000021934509},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.46860000491142273},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.43869999051094055},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.4138999879360199},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.4120999872684479},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.40639999508857727},{"id":"https://openalex.org/C193969084","wikidata":"https://www.wikidata.org/wiki/Q7452500","display_name":"Sequential decoding","level":4,"score":0.388700008392334},{"id":"https://openalex.org/C75291252","wikidata":"https://www.wikidata.org/wiki/Q1315756","display_name":"TRACE (psycholinguistics)","level":2,"score":0.3529999852180481},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.3490000069141388},{"id":"https://openalex.org/C2778029271","wikidata":"https://www.wikidata.org/wiki/Q5421931","display_name":"Extension (predicate logic)","level":2,"score":0.3400000035762787},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3239000141620636},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.3163999915122986},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.31540000438690186},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.3018999993801117},{"id":"https://openalex.org/C45235069","wikidata":"https://www.wikidata.org/wiki/Q278425","display_name":"Table (database)","level":2,"score":0.2971999943256378},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.2870999872684479},{"id":"https://openalex.org/C204397858","wikidata":"https://www.wikidata.org/wiki/Q4437907","display_name":"List decoding","level":5,"score":0.2856999933719635},{"id":"https://openalex.org/C2781039887","wikidata":"https://www.wikidata.org/wiki/Q1391724","display_name":"Factor (programming language)","level":2,"score":0.2842000126838684},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.2816999852657318},{"id":"https://openalex.org/C99138194","wikidata":"https://www.wikidata.org/wiki/Q183427","display_name":"Hash function","level":2,"score":0.27469998598098755},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.274399995803833},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C101454708","wikidata":"https://www.wikidata.org/wiki/Q17106019","display_name":"Standard Model (mathematical formulation)","level":3,"score":0.262800008058548},{"id":"https://openalex.org/C97256817","wikidata":"https://www.wikidata.org/wiki/Q1462316","display_name":"Spurious relationship","level":2,"score":0.2524000108242035},{"id":"https://openalex.org/C66882249","wikidata":"https://www.wikidata.org/wiki/Q169336","display_name":"Homogeneous","level":2,"score":0.25119999051094055}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.02955","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.02955","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.2606.02955","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.02955","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":{"Diffusion":[0],"large":[1],"language":[2],"models":[3],"promise":[4],"parallel":[5,30,81],"token":[6],"generation,":[7],"yet":[8],"inference":[9],"remains":[10],"bottlenecked":[11],"by":[12,178],"deciding":[13],"which":[14],"masked":[15],"tokens":[16,127],"can":[17],"be":[18],"safely":[19],"committed":[20],"together.":[21],"Fast-dLLM":[22,149],"addressed":[23],"this":[24,55],"with":[25,158],"KV":[26],"caching":[27],"and":[28,107,118,137,156],"confidence-guided":[29],"decoding,":[31],"but":[32],"its":[33,48],"decoding":[34,63],"theory":[35],"uses":[36],"a":[37,72,92,100,120,144],"homogeneous":[38],"high-confidence":[39],"assumption":[40],"that":[41,54,75,163,182],"effectively":[42],"reduces":[43],"each":[44],"candidate":[45],"set":[46],"to":[47,188],"weakest":[49],"selected":[50,126],"token.":[51],"We":[52,69],"argue":[53],"leaves":[56,132],"speed":[57],"on":[58,152],"the":[59,85,110,115,125,133,159,164,175],"table":[60],"because":[61],"real":[62],"steps":[64],"exhibit":[65],"heterogeneous":[66],"confidence":[67,88],"profiles.":[68],"propose":[70],"\\textbf{Fast-dLLM++},":[71],"training-free":[73],"extension":[74],"introduces":[76],"\\emph{Fr\u00e9chet":[77],"profile":[78,89],"decoding}:":[79],"selecting":[80],"commit":[82],"sets":[83],"from":[84],"full":[86],"sorted":[87],"rather":[90],"than":[91],"single":[93],"worst-case":[94],"confidence.":[95],"The":[96],"resulting":[97],"rule":[98,112],"is":[99,198],"heterogeneous-confidence":[101],"generalization":[102],"of":[103],"Fast-dLLM's":[104],"factor":[105],"selector":[106],"it":[108,143],"recovers":[109],"previous":[111],"exactly":[113],"in":[114],"equal-confidence":[116],"case":[117],"adds":[119],"provable":[121],"\\emph{heterogeneity":[122],"bonus}":[123],"when":[124],"have":[128],"uneven":[129],"confidences.":[130],"Fast-dLLM++":[131],"model,":[134],"diffusion":[135],"process,":[136],"cache":[138],"implementation":[139],"entirely":[140],"unchanged,":[141],"making":[142],"drop-in":[145],"replacement":[146],"for":[147],"existing":[148],"decoding.":[150],"Experiments":[151],"GSM8K,":[153],"MATH,":[154],"HumanEval,":[155],"MBPP":[157],"LLaDA-8B":[160],"model":[161],"show":[162],"theoretical":[165],"improvement":[166],"translates":[167],"directly":[168],"into":[169],"empirical":[170],"gains:":[171],"profile-aware":[172],"selection":[173],"improves":[174],"accuracy--throughput":[176],"frontier":[177],"exploiting":[179],"safe":[180],"parallelism":[181],"weakest-token":[183],"rules":[184],"miss,":[185],"achieving":[186],"up":[187],"37\\%":[189],"higher":[190],"throughput":[191],"at":[192,199],"comparable":[193],"accuracy.":[194],"Our":[195],"code":[196],"release":[197],"https://github.com/Ringo-Star/FastdLLM_plusplus.":[200]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-04T00:00:00"}