{"id":"https://openalex.org/W7161178916","doi":"https://doi.org/10.48550/arxiv.2605.13316","title":"Test-time Sparsity for Extreme Fast Action Diffusion","display_name":"Test-time Sparsity for Extreme Fast Action Diffusion","publication_year":2026,"publication_date":"2026-05-13","ids":{"openalex":"https://openalex.org/W7161178916","doi":"https://doi.org/10.48550/arxiv.2605.13316"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.13316","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.13316","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.2605.13316","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136152278","display_name":"Kangye Ji","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ji, Kangye","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136103262","display_name":"Yuan Meng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Meng, Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023231504","display_name":"Jianbo Zhou","orcid":"https://orcid.org/0000-0002-5432-0945"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Jianbo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136105945","display_name":"Ye Li","orcid":"https://orcid.org/0000-0002-5351-8546"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Ye","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136101351","display_name":"Chen Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136177315","display_name":"Zhi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zhi","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":0.6255999803543091,"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":0.6255999803543091,"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/T11349","display_name":"Music Technology and Sound Studies","score":0.043699998408555984,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.041200000792741776,"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/noise-reduction","display_name":"Noise reduction","score":0.5145999789237976},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.5102999806404114},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4555000066757202},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.43650001287460327},{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.42309999465942383},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.4101000130176544},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.3953000009059906},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.37770000100135803},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.36739999055862427}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6970000267028809},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.5145999789237976},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.5102999806404114},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4997999966144562},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4555000066757202},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.43650001287460327},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.42309999465942383},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.4101000130176544},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.3953000009059906},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.37770000100135803},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.36739999055862427},{"id":"https://openalex.org/C2781137444","wikidata":"https://www.wikidata.org/wiki/Q237105","display_name":"Sharpening","level":2,"score":0.3521000146865845},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.3499999940395355},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34139999747276306},{"id":"https://openalex.org/C187691185","wikidata":"https://www.wikidata.org/wiki/Q2020720","display_name":"Grid","level":2,"score":0.3398999869823456},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.3237999975681305},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.29330000281333923},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.29089999198913574},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.2883000075817108},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.2865999937057495},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.2721000015735626},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.263700008392334},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.2558000087738037},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.25369998812675476},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.25270000100135803},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.25220000743865967}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.13316","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.13316","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.2605.13316","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.13316","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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Action":[0],"diffusion":[1,25,63,144,223],"excels":[2],"at":[3,74,257],"high-fidelity":[4],"action":[5,62,217,239],"generation":[6,240],"but":[7],"incurs":[8],"heavy":[9],"computational":[10],"costs":[11],"owing":[12],"to":[13,34,36,54,60,129,219],"its":[14],"iterative":[15],"denoising":[16,102,156,161,201],"nature.":[17],"Despite":[18],"current":[19,198],"technologies":[20],"showing":[21],"promise":[22],"in":[23,47,81,163],"accelerating":[24],"transformers":[26],"by":[27,64,158,191,225,235,241],"reusing":[28,185,193,211],"the":[29,97,114,126,140,143,149,154,167,170,178,197,209,221],"cached":[30,99,195],"features,":[31],"they":[32],"struggle":[33],"adapt":[35],"policy":[37],"dynamics":[38],"arising":[39],"from":[40,100,153,196],"diverse":[41],"perceptions":[42],"and":[43,88,95,151,165,203,237],"multi-round":[44],"rollout":[45,205],"iterations":[46],"open":[48],"environments.":[49],"We":[50],"propose":[51],"test-time":[52],"sparsity":[53,190],"tackle":[55],"this":[56,82],"challenge,":[57],"which":[58,187],"aims":[59],"accelerate":[61],"dynamically":[65],"predicting":[66],"prunable":[67],"residual":[68],"computations":[69],"for":[70],"each":[71],"model":[72],"forward":[73,172],"test":[75],"time.":[76],"However,":[77],"two":[78],"bottlenecks":[79],"remain":[80],"paradigm:":[83],"1)":[84],"repetitive":[85],"conditional":[86],"encoding":[87,150],"pruning":[89,107,152],"offset":[90],"most":[91],"potential":[92],"speed":[93],"gains,":[94],"2)":[96],"features":[98,194],"previous":[101,200],"timesteps":[103,162],"cannot":[104],"constrain":[105],"large":[106],"errors":[108],"under":[109],"aggressive":[110],"sparsity.":[111],"To":[112,176,207],"address":[113],"first":[115,133],"bottleneck,":[116,180],"we":[117,132,147,181,213],"design":[118,134],"a":[119,135,215],"highly":[120],"parallelized":[121],"inference":[122,173,248],"pipeline":[123],"that":[124,138,230],"minimizes":[125],"non-decoder":[127],"delay":[128],"milliseconds.":[130],"Specifically,":[131],"lightweight":[136],"pruner":[137,168],"shares":[139],"encoder":[141],"with":[142,169,246],"transformer.":[145],"Then,":[146],"decouple":[148],"autoregressive":[155],"loop":[157],"processing":[159],"all":[160],"parallel,":[164],"overlap":[166],"decoder":[171],"through":[174],"asynchronism.":[175],"overcome":[177],"second":[179],"introduce":[182],"an":[183,247],"omnidirectional":[184],"strategy,":[186],"achieves":[188],"95%":[189],"selectively":[192],"forward,":[199],"timesteps,":[202],"earlier":[204],"iterations.":[206],"learn":[208],"rollout-level":[210],"strategies,":[212],"sample":[214],"few":[216],"trajectories":[218],"supervise":[220],"sparsified":[222],"step":[224],"step.":[226],"Extensive":[227],"experiments":[228],"demonstrate":[229],"our":[231],"method":[232],"reduces":[233],"FLOPs":[234],"92%":[236],"accelerates":[238],"5x,":[242],"achieving":[243],"lossless":[244],"performance":[245],"frequency":[249],"of":[250],"47.5":[251],"Hz.":[252],"Our":[253],"code":[254],"is":[255],"available":[256],"https://github.com/ky-ji/Test-time-Sparsity.":[258]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-15T00:00:00"}