{"id":"https://openalex.org/W4415329583","doi":"https://doi.org/10.48550/arxiv.2505.16474","title":"Foresight Diffusion: Improving Sampling Consistency in Predictive Diffusion Models","display_name":"Foresight Diffusion: Improving Sampling Consistency in Predictive Diffusion Models","publication_year":2025,"publication_date":"2025-05-22","ids":{"openalex":"https://openalex.org/W4415329583","doi":"https://doi.org/10.48550/arxiv.2505.16474"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2505.16474","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.16474","pdf_url":"https://arxiv.org/pdf/2505.16474","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2505.16474","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100433704","display_name":"Yu Zhang","orcid":"https://orcid.org/0000-0003-1836-645X","observed_orcids":["https://orcid.org/0000-0003-1836-645X"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhang, Yu","raw_affiliation_strings":["School of Software , BNRist , Tsinghua University , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Software , BNRist , Tsinghua University , China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018545534","display_name":"Xingzhuo Guo","orcid":"https://orcid.org/0009-0006-8968-5633","observed_orcids":["https://orcid.org/0009-0006-8968-5633"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guo, Xingzhuo","raw_affiliation_strings":["School of Software , BNRist , Tsinghua University , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Software , BNRist , Tsinghua University , China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103168380","display_name":"Haoran Xu","orcid":"https://orcid.org/0000-0002-0213-207X","observed_orcids":["https://orcid.org/0000-0002-0213-207X"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xu, Haoran","raw_affiliation_strings":["School of Software , BNRist , Tsinghua University , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Software , BNRist , Tsinghua University , China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Wu, Jialong","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Jialong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5019241553","display_name":"Mingsheng Long","orcid":"https://orcid.org/0000-0001-9421-463X","observed_orcids":["https://orcid.org/0000-0001-9421-463X"]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Long, Mingsheng","raw_affiliation_strings":["School of Software , BNRist , Tsinghua University , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Software , BNRist , Tsinghua University , China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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":0.9761000275611877,"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":0.9761000275611877,"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/T11206","display_name":"Model Reduction and Neural Networks","score":0.010200000368058681,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.003000000026077032,"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/spatiotemporal-prediction","display_name":"spatiotemporal prediction","score":0.9380000233650208},{"id":"https://openalex.org/keywords/conditional-denoising","display_name":"conditional denoising","score":0.43799999356269836}],"concepts":[{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.5799999833106995},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5788999795913696},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5716000199317932},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.5551999807357788},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4896000027656555},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.48019999265670776},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.41679999232292175},{"id":"https://openalex.org/C205606062","wikidata":"https://www.wikidata.org/wiki/Q5249645","display_name":"Decoupling (probability)","level":2,"score":0.37290000915527344},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.36070001125335693},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.35600000619888306},{"id":"https://openalex.org/C64848388","wikidata":"https://www.wikidata.org/wiki/Q188867","display_name":"Futures studies","level":2,"score":0.3441999852657318},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.33500000834465027},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.3075999915599823},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.30379998683929443},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.2946999967098236},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.2840999960899353},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2741999924182892},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.2590999901294708},{"id":"https://openalex.org/C68710425","wikidata":"https://www.wikidata.org/wiki/Q5275442","display_name":"Diffusion process","level":3,"score":0.25029999017715454}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2505.16474","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.16474","pdf_url":"https://arxiv.org/pdf/2505.16474","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2505.16474","doi":"https://doi.org/10.48550/arxiv.2505.16474","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2505.16474","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"pmh:oai:arXiv.org:2505.16474","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.16474","pdf_url":"https://arxiv.org/pdf/2505.16474","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4415329583.pdf","grobid_xml":"https://content.openalex.org/works/W4415329583.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Diffusion":[0,98],"and":[1,14,38,83,89,133,152,163],"flow-based":[2],"models":[3,68,105],"have":[4,15],"enabled":[5],"significant":[6],"progress":[7],"in":[8,19,54,63,70],"generation":[9,25],"tasks":[10,26],"across":[11],"various":[12],"modalities":[13],"recently":[16],"found":[17],"applications":[18],"predictive":[20,31,66,72,103,122,161,174],"learning.":[21],"However,":[22],"unlike":[23],"typical":[24],"that":[27,59,106,143,157],"encourage":[28],"sample":[29],"diversity,":[30],"learning":[32,64],"entails":[33],"different":[34],"sources":[35],"of":[36,80,129],"stochasticity":[37],"requires":[39],"sampling":[40,108,164],"consistency":[41,109,165],"aligned":[42],"with":[43],"the":[44,78,130],"ground-truth":[45],"trajectory,":[46],"which":[47,74],"is":[48],"a":[49,60,100,119,136,170],"limitation":[50],"we":[51,75,95],"empirically":[52],"observe":[53],"diffusion":[55,67,104,175],"models.":[56,176],"We":[57],"argue":[58],"key":[61],"bottleneck":[62],"sampling-consistent":[65],"lies":[69],"suboptimal":[71],"ability,":[73],"attribute":[76],"to":[77,124,139],"entanglement":[79],"condition":[81,112],"understanding":[82,113],"target":[84,115],"denoising":[85,131],"within":[86],"shared":[87],"architectures":[88],"co-training":[90],"schemes.":[91],"To":[92],"address":[93],"this,":[94],"propose":[96],"Foresight":[97],"(ForeDiff),":[99],"framework":[101],"for":[102,173],"improves":[107,159],"by":[110],"decoupling":[111],"from":[114],"denoising.":[116],"ForeDiff":[117,158],"incorporates":[118],"separate":[120],"deterministic":[121],"stream":[123],"process":[125],"conditioning":[126],"inputs":[127],"independently":[128],"stream,":[132],"further":[134],"leverages":[135],"pretrained":[137],"predictor":[138],"extract":[140],"informative":[141],"representations":[142],"guide":[144],"generation.":[145],"Extensive":[146],"experiments":[147],"on":[148],"robot":[149],"video":[150],"prediction":[151],"scientific":[153],"spatiotemporal":[154],"forecasting":[155],"show":[156],"both":[160],"accuracy":[162],"over":[166],"strong":[167],"baselines,":[168],"offering":[169],"promising":[171],"direction":[172]},"counts_by_year":[],"updated_date":"2026-09-30T09:27:27.532063","created_date":"2025-10-19T00:00:00"}