{"id":"https://openalex.org/W7155377970","doi":"https://doi.org/10.48550/arxiv.2604.20289","title":"X-Cache: Cross-Chunk Block Caching for Few-Step Autoregressive World Models Inference","display_name":"X-Cache: Cross-Chunk Block Caching for Few-Step Autoregressive World Models Inference","publication_year":2026,"publication_date":"2026-04-22","ids":{"openalex":"https://openalex.org/W7155377970","doi":"https://doi.org/10.48550/arxiv.2604.20289"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.20289","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.20289","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.20289","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134384601","display_name":"Yixiao Zeng","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Zeng, Yixiao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134389497","display_name":"Jianlei Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Jianlei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015757467","display_name":"Chaoda Zheng","orcid":"https://orcid.org/0000-0001-9347-4181"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Chaoda","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130335202","display_name":"Shijia Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Shijia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134378970","display_name":"Mingdian Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Mingdian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134374394","display_name":"Tongping Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Tongping","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134400151","display_name":"Tengwei Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luo, Tengwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134464032","display_name":"Yu Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134404061","display_name":"Boyang Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Boyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134380162","display_name":"Linkun Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Linkun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134404395","display_name":"Siyuan Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Siyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101610890","display_name":"Bo Tian","orcid":"https://orcid.org/0000-0002-9722-2528"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tian, Bo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5134456550","display_name":"Xianming Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Xianming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":13,"corresponding_author_ids":["https://openalex.org/A5134384601"],"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":0.5471000075340271,"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":0.5471000075340271,"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/T12290","display_name":"Human Motion and Animation","score":0.0778999999165535,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.04039999842643738,"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/autoregressive-model","display_name":"Autoregressive model","score":0.5964999794960022},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.5382000207901001},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4724999964237213},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.4724000096321106},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.4440000057220459},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.40799999237060547},{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.40070000290870667},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.38600000739097595}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7799000144004822},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.5964999794960022},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.5382000207901001},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4724999964237213},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.4724000096321106},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.4440000057220459},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.40799999237060547},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.40070000290870667},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.38600000739097595},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.37860000133514404},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.3774999976158142},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.3610999882221222},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.35899999737739563},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.351500004529953},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.3345000147819519},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.32359999418258667},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.31630000472068787},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.30820000171661377},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.29260000586509705},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.2919999957084656},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.2896000146865845},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2890999913215637},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.2680000066757202},{"id":"https://openalex.org/C187691185","wikidata":"https://www.wikidata.org/wiki/Q2020720","display_name":"Grid","level":2,"score":0.25540000200271606},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.25209999084472656}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.20289","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.20289","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.20289","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.20289","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":"article"},"sustainable_development_goals":[{"display_name":"Industry, innovation and infrastructure","score":0.6554522514343262,"id":"https://metadata.un.org/sdg/9"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Real-time":[0],"world":[1,21,200],"simulation":[2],"is":[3],"becoming":[4],"a":[5,38,95,102,125,130,195],"key":[6],"infrastructure":[7],"for":[8,40,50,73],"scalable":[9],"evaluation":[10],"and":[11,58,78,123,132,172,178,210],"online":[12],"reinforcement":[13],"learning":[14],"of":[15],"autonomous":[16],"driving":[17,20,199],"systems.":[18],"Recent":[19],"models":[22,67],"built":[23,202],"on":[24,183,193,203],"autoregressive":[25,157],"video":[26,52],"diffusion":[27,45],"achieve":[28],"high-fidelity,":[29],"controllable":[30],"multi-camera":[31,197],"generation,":[32],"but":[33],"their":[34],"inference":[35],"cost":[36],"remains":[37],"bottleneck":[39],"interactive":[41,87],"deployment.":[42],"However,":[43],"existing":[44],"caching":[46],"methods":[47,75],"are":[48],"designed":[49],"offline":[51],"generation":[53,88,107],"with":[54,207,220],"multiple":[55],"denoising":[56,112,209],"steps,":[57],"do":[59],"not":[60,90],"transfer":[61],"to":[62,76,136],"this":[63],"scenario.":[64],"Few-step":[65],"distilled":[66],"have":[68],"no":[69],"inter-step":[70],"redundancy":[71],"left":[72],"these":[74,184],"reuse,":[77],"sequence-level":[79],"parallelization":[80],"techniques":[81],"require":[82],"future":[83],"conditioning":[84],"that":[85,99,119,168],"closed-loop":[86],"does":[89],"provide.":[91],"We":[92,190],"present":[93],"X-Cache,":[94],"training-free":[96],"acceleration":[97],"method":[98],"caches":[100,118],"along":[101],"different":[103],"axis:":[104],"across":[105,111,121],"consecutive":[106],"chunks":[108,164],"rather":[109],"than":[110],"steps.":[113],"X-Cache":[114,160,192,214],"maintains":[115],"per-block":[116],"residual":[117],"persist":[120],"chunks,":[122,185],"applies":[124],"dual-metric":[126],"gating":[127],"mechanism":[128],"over":[129],"structure-":[131],"action-aware":[133],"block-input":[134],"fingerprint":[135],"independently":[137],"decide":[138],"whether":[139],"each":[140],"block":[141,217],"should":[142],"recompute":[143],"or":[144],"reuse":[145],"its":[146],"cached":[147],"residual.":[148],"To":[149],"prevent":[150],"approximation":[151],"errors":[152],"from":[153],"permanently":[154],"contaminating":[155],"the":[156,175],"KV":[158,162,212],"cache,":[159],"identifies":[161],"update":[163],"(the":[165],"forward":[166],"passes":[167],"write":[169],"clean":[170],"keys":[171],"values":[173],"into":[174],"persistent":[176],"cache)":[177],"unconditionally":[179],"forces":[180],"full":[181],"computation":[182],"cutting":[186],"off":[187],"error":[188],"propagation.":[189],"implement":[191],"X-world,":[194],"production":[196],"action-conditioned":[198],"model":[201],"multi-block":[204],"causal":[205],"DiT":[206],"few-step":[208],"rolling":[211],"cache.":[213],"achieves":[215],"71%":[216],"skip":[218],"rate":[219],"2.6x":[221],"wall-clock":[222],"speedup":[223],"while":[224],"maintaining":[225],"minimum":[226],"degradation.":[227]},"counts_by_year":[],"updated_date":"2026-04-24T06:07:52.864757","created_date":"2026-04-24T00:00:00"}