{"id":"https://openalex.org/W7163702610","doi":"https://doi.org/10.48550/arxiv.2606.05742","title":"AdaPLD: Adaptive Retrieval and Reuse for Efficient Model-Free Speculative Decoding","display_name":"AdaPLD: Adaptive Retrieval and Reuse for Efficient Model-Free Speculative Decoding","publication_year":2026,"publication_date":"2026-06-04","ids":{"openalex":"https://openalex.org/W7163702610","doi":"https://doi.org/10.48550/arxiv.2606.05742"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.05742","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.05742","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.2606.05742","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137934322","display_name":"Runheng Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Runheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137954885","display_name":"Jincheng Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Jincheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138002418","display_name":"Wen Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Wen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137924675","display_name":"Xingchen Xiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao, Xingchen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137953862","display_name":"Heyan Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Heyan","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":0.3824999928474426,"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":0.3824999928474426,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.19030000269412994,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.11670000106096268,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.7368000149726868},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.6834999918937683},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5450000166893005},{"id":"https://openalex.org/keywords/copying","display_name":"Copying","score":0.4587000012397766},{"id":"https://openalex.org/keywords/similarity","display_name":"Similarity (geometry)","score":0.39480000734329224},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.37540000677108765},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.3490000069141388},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.34439998865127563}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7770000100135803},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.7368000149726868},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.6834999918937683},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5450000166893005},{"id":"https://openalex.org/C2779151265","wikidata":"https://www.wikidata.org/wiki/Q1156791","display_name":"Copying","level":2,"score":0.4587000012397766},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.39480000734329224},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.37540000677108765},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.3490000069141388},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.34439998865127563},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.34049999713897705},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3285999894142151},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3098999857902527},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.30160000920295715},{"id":"https://openalex.org/C81669768","wikidata":"https://www.wikidata.org/wiki/Q2359161","display_name":"Precision and recall","level":2,"score":0.3001999855041504},{"id":"https://openalex.org/C48105269","wikidata":"https://www.wikidata.org/wiki/Q1141160","display_name":"Header","level":2,"score":0.295199990272522},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2775999903678894},{"id":"https://openalex.org/C186644900","wikidata":"https://www.wikidata.org/wiki/Q194152","display_name":"Parsing","level":2,"score":0.2630999982357025},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2581000030040741},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.25369998812675476},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2533000111579895},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.2508000135421753}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.05742","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.05742","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.2606.05742","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.05742","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":[1,17,144],"accelerates":[2],"generation":[3],"by":[4,25],"verifying":[5],"multiple":[6],"drafted":[7],"tokens":[8],"in":[9],"a":[10,83,127],"single":[11,128],"target-model":[12,136],"forward":[13,137],"pass,":[14],"reducing":[15],"sequential":[16],"iterations.":[18],"Model-free":[19],"variants":[20],"avoid":[21],"auxiliary":[22],"draft":[23,92],"models":[24],"reusing":[26],"text":[27],"and":[28,63,91,139],"model":[29],"states":[30],"already":[31],"available":[32],"during":[33],"generation,":[34],"but":[35],"their":[36],"speedup":[37],"depends":[38],"on":[39,126],"the":[40,43,71,78],"reliability":[41],"of":[42,50],"constructed":[44],"drafts.":[45],"We":[46,80],"identify":[47],"two":[48],"limitations":[49],"existing":[51],"reuse-based":[52],"methods:":[53],"lexically":[54],"anchored":[55],"retrieval":[56,90],"has":[57],"limited":[58],"recall":[59],"under":[60],"surface-form":[61],"variation,":[62],"deterministic":[64],"span":[65],"copying":[66],"can":[67],"be":[68],"brittle":[69],"when":[70,108],"retrieved":[72],"context":[73],"does":[74],"not":[75],"uniquely":[76],"determine":[77],"continuation.":[79],"propose":[81],"\\emph{AdaPLD},":[82],"training-free":[84],"method":[85],"that":[86],"adaptively":[87],"improves":[88],"both":[89],"construction.":[93],"AdaPLD":[94,134],"preserves":[95],"high-precision":[96],"lexical":[97,109],"reuse":[98,106,116],"while":[99],"using":[100],"semantic":[101],"similarity":[102],"to":[103,118,142],"recover":[104],"additional":[105],"opportunities":[107],"matching":[110],"fails.":[111],"It":[112],"further":[113],"constructs":[114],"branched":[115],"hypotheses":[117],"account":[119],"for":[120],"continuation":[121],"uncertainty,":[122],"rather":[123],"than":[124],"relying":[125],"copied":[129],"span.":[130],"Across":[131],"diverse":[132],"benchmarks,":[133],"reduces":[135],"passes":[138],"achieves":[140],"up":[141],"$3.10\\times$":[143],"speedup.":[145]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-06T00:00:00"}