{"id":"https://openalex.org/W4417169537","doi":"https://doi.org/10.1109/tpami.2025.3641798","title":"Toward Accurate Procedure Planning in Instructional Videos: Visual State Generation Helps Task-Selective Diffusion","display_name":"Toward Accurate Procedure Planning in Instructional Videos: Visual State Generation Helps Task-Selective Diffusion","publication_year":2025,"publication_date":"2025-12-09","ids":{"openalex":"https://openalex.org/W4417169537","doi":"https://doi.org/10.1109/tpami.2025.3641798","pmid":"https://pubmed.ncbi.nlm.nih.gov/41364568"},"language":"en","primary_location":{"id":"doi:10.1109/tpami.2025.3641798","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2025.3641798","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},"type":"article","indexed_in":["crossref","pubmed"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5030374805","display_name":"Fen Fang","orcid":"https://orcid.org/0000-0002-3834-4795"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3005327000","display_name":"Institute for Infocomm Research","ror":"https://ror.org/053rfa017","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3005327000","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Fen Fang","raw_affiliation_strings":["Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore"],"raw_orcid":"https://orcid.org/0000-0002-3834-4795","affiliations":[{"raw_affiliation_string":"Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]},{"raw_affiliation_string":"Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020925728","display_name":"Muli Yang","orcid":"https://orcid.org/0000-0002-5959-2931"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3005327000","display_name":"Institute for Infocomm Research","ror":"https://ror.org/053rfa017","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3005327000","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Muli Yang","raw_affiliation_strings":["Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore"],"raw_orcid":"https://orcid.org/0000-0002-5959-2931","affiliations":[{"raw_affiliation_string":"Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]},{"raw_affiliation_string":"Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100340964","display_name":"Min Wu","orcid":"https://orcid.org/0000-0003-0977-3600"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3005327000","display_name":"Institute for Infocomm Research","ror":"https://ror.org/053rfa017","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3005327000","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Min Wu","raw_affiliation_strings":["Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore"],"raw_orcid":"https://orcid.org/0000-0003-0977-3600","affiliations":[{"raw_affiliation_string":"Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]},{"raw_affiliation_string":"Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035747364","display_name":"Yanhua Yang","orcid":"https://orcid.org/0000-0002-7916-3683"},"institutions":[{"id":"https://openalex.org/I149594827","display_name":"Xidian University","ror":"https://ror.org/05s92vm98","country_code":"CN","type":"education","lineage":["https://openalex.org/I149594827"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanhua Yang","raw_affiliation_strings":["School of Computer Science and Technology, Xidian University, Xi&#x2019;an, China","School of Computer Science and Technology, Xidian University, Xi&#x0027;an, China"],"raw_orcid":"https://orcid.org/0000-0002-7916-3683","affiliations":[{"raw_affiliation_string":"School of Computer Science and Technology, Xidian University, Xi&#x2019;an, China","institution_ids":["https://openalex.org/I149594827"]},{"raw_affiliation_string":"School of Computer Science and Technology, Xidian University, Xi&#x0027;an, China","institution_ids":["https://openalex.org/I149594827"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052143634","display_name":"Qianli Xu","orcid":"https://orcid.org/0000-0003-0105-5903"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3005327000","display_name":"Institute for Infocomm Research","ror":"https://ror.org/053rfa017","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3005327000","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Qianli Xu","raw_affiliation_strings":["Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore"],"raw_orcid":"https://orcid.org/0000-0003-0105-5903","affiliations":[{"raw_affiliation_string":"Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]},{"raw_affiliation_string":"Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077258552","display_name":"Joo\u2010Hwee Lim","orcid":"https://orcid.org/0000-0002-4103-3824"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3005327000","display_name":"Institute for Infocomm Research","ror":"https://ror.org/053rfa017","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3005327000","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Joo-Hwee Lim","raw_affiliation_strings":["Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore"],"raw_orcid":"https://orcid.org/0000-0002-4103-3824","affiliations":[{"raw_affiliation_string":"Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]},{"raw_affiliation_string":"Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029757365","display_name":"Xulei Yang","orcid":"https://orcid.org/0000-0002-7002-4564"},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3005327000","display_name":"Institute for Infocomm Research","ror":"https://ror.org/053rfa017","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3005327000","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Xulei Yang","raw_affiliation_strings":["Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore"],"raw_orcid":"https://orcid.org/0000-0002-7002-4564","affiliations":[{"raw_affiliation_string":"Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]},{"raw_affiliation_string":"Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110463782","display_name":"Hongyuan Zhu","orcid":null},"institutions":[{"id":"https://openalex.org/I115228651","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09","country_code":"SG","type":"government","lineage":["https://openalex.org/I115228651"]},{"id":"https://openalex.org/I3005327000","display_name":"Institute for Infocomm Research","ror":"https://ror.org/053rfa017","country_code":"SG","type":"facility","lineage":["https://openalex.org/I115228651","https://openalex.org/I3005327000","https://openalex.org/I91275662"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Hongyuan Zhu","raw_affiliation_strings":["Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute for Infocomm Research (I<sup>2</sup>R), Agency for Science, Technology and Research (A&#x002A;STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]},{"raw_affiliation_string":"Institute for Infocomm Research (IR), Agency for Science, Technology and Research (A*STAR), Singapore","institution_ids":["https://openalex.org/I3005327000","https://openalex.org/I115228651"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.36143242,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"48","issue":"4","first_page":"4033","last_page":"4050"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.855400025844574,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.855400025844574,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.03350000083446503,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.027300000190734863,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6847000122070312},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.6771000027656555},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.593999981880188},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5687000155448914},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5426999926567078},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.5285000205039978},{"id":"https://openalex.org/keywords/plan","display_name":"Plan (archaeology)","score":0.4805999994277954},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.450300008058548}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8360999822616577},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6847000122070312},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6771000027656555},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6305000185966492},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.593999981880188},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5687000155448914},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5426999926567078},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5357999801635742},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.5285000205039978},{"id":"https://openalex.org/C2776505523","wikidata":"https://www.wikidata.org/wiki/Q4785468","display_name":"Plan (archaeology)","level":2,"score":0.4805999994277954},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.450300008058548},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.41190001368522644},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.3912999927997589},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.3905999958515167},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.3643999993801117},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.35280001163482666},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3490000069141388},{"id":"https://openalex.org/C166109690","wikidata":"https://www.wikidata.org/wiki/Q4677422","display_name":"Action selection","level":3,"score":0.26910001039505005},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.26899999380111694},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.25380000472068787}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/tpami.2025.3641798","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tpami.2025.3641798","pdf_url":null,"source":{"id":"https://openalex.org/S199944782","display_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","issn_l":"0162-8828","issn":["0162-8828","1939-3539","2160-9292"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Pattern Analysis and Machine Intelligence","raw_type":"journal-article"},{"id":"pmid:41364568","is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/41364568","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE transactions on pattern analysis and machine intelligence","raw_type":null}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6103525054","display_name":null,"funder_award_id":"62571412","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320320696","display_name":"Agency for Science, Technology and Research","ror":"https://ror.org/036wvzt09"},{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":73,"referenced_works":["https://openalex.org/W1901129140","https://openalex.org/W2185953016","https://openalex.org/W2422305492","https://openalex.org/W2528489519","https://openalex.org/W2550143307","https://openalex.org/W2592749667","https://openalex.org/W2749658623","https://openalex.org/W2777985721","https://openalex.org/W2798708692","https://openalex.org/W2945792291","https://openalex.org/W2951098185","https://openalex.org/W2952132648","https://openalex.org/W2957775769","https://openalex.org/W2962795934","https://openalex.org/W2963302737","https://openalex.org/W2963570630","https://openalex.org/W2963634205","https://openalex.org/W2963642716","https://openalex.org/W2964094654","https://openalex.org/W2979490629","https://openalex.org/W2981635073","https://openalex.org/W2981851019","https://openalex.org/W2984008963","https://openalex.org/W2993447238","https://openalex.org/W3035467150","https://openalex.org/W3035635319","https://openalex.org/W3094502228","https://openalex.org/W3106697459","https://openalex.org/W3106768499","https://openalex.org/W3108908812","https://openalex.org/W3177173029","https://openalex.org/W3199332348","https://openalex.org/W3203256294","https://openalex.org/W3203711169","https://openalex.org/W3205786327","https://openalex.org/W3207758636","https://openalex.org/W3211485050","https://openalex.org/W4221153256","https://openalex.org/W4307845887","https://openalex.org/W4312388283","https://openalex.org/W4312478760","https://openalex.org/W4312518915","https://openalex.org/W4312909532","https://openalex.org/W4312919100","https://openalex.org/W4312933868","https://openalex.org/W4313023218","https://openalex.org/W4313038449","https://openalex.org/W4313055276","https://openalex.org/W4313118773","https://openalex.org/W4313174527","https://openalex.org/W4322764164","https://openalex.org/W4386065447","https://openalex.org/W4386065648","https://openalex.org/W4386065689","https://openalex.org/W4386065873","https://openalex.org/W4386075514","https://openalex.org/W4386075592","https://openalex.org/W4386075710","https://openalex.org/W4386076314","https://openalex.org/W4386076369","https://openalex.org/W4390872096","https://openalex.org/W4390872546","https://openalex.org/W4390873035","https://openalex.org/W4390874188","https://openalex.org/W4390874412","https://openalex.org/W4394699010","https://openalex.org/W4402770916","https://openalex.org/W4403842924","https://openalex.org/W4405440310","https://openalex.org/W6922480057","https://openalex.org/W7133185348","https://openalex.org/W7133193597","https://openalex.org/W7133204208"],"related_works":[],"abstract_inverted_index":{"Procedure":[0],"planning":[1],"in":[2,44],"instructional":[3],"videos":[4],"entails":[5],"predicting":[6],"an":[7,37,80],"action":[8,133],"sequence":[9],"that":[10,73,125],"transitions":[11],"a":[12,17,64,105,111,121,127],"given":[13],"start":[14],"state":[15],"to":[16,27,50,93,130,156],"desired":[18],"goal":[19],"state.":[20],"This":[21],"task":[22,144],"is":[23],"particularly":[24],"challenging":[25],"due":[26,49],"two":[28],"key":[29],"sources":[30],"of":[31,67,138,172],"uncertainty:":[32],"limited":[33],"visual":[34,53,97,149],"observations":[35],"and":[36],"enormous":[38],"decision":[39],"space.":[40,134],"The":[41],"former":[42],"results":[43],"multiple":[45],"plausible":[46],"plan":[47],"variations":[48],"missing":[51],"intermediate":[52,96],"states,":[54],"while":[55],"the":[56,85,116,132,136,169],"latter":[57],"complicates":[58],"prediction":[59],"by":[60,104,151],"requiring":[61],"selection":[62,107],"from":[63],"large":[65],"set":[66],"potential":[68],"actions.":[69],"Unlike":[70],"prior":[71],"work":[72],"addresses":[74],"these":[75],"issues":[76],"implicitly,":[77],"we":[78,88,119,146],"propose":[79],"explicit":[81],"solution.":[82],"To":[83,114],"mitigate":[84],"first":[86],"challenge,":[87,118],"employ":[89],"image":[90],"generation":[91],"models":[92,155],"synthesize":[94],"diverse":[95],"states":[98],"using":[99],"various":[100],"text":[101],"prompts,":[102],"followed":[103],"prompt":[106],"module":[108],"integrated":[109],"within":[110],"diffusion":[112,123],"model.":[113],"tackle":[115],"second":[117],"introduce":[120],"task-selective":[122],"model":[124],"applies":[126],"task-specific":[128],"mask":[129,140],"constrain":[131],"As":[135],"effectiveness":[137],"this":[139],"depends":[141],"on":[142,164],"accurate":[143],"classification,":[145],"further":[147],"enhance":[148],"representation":[150],"leveraging":[152],"pre-trained":[153],"vision-language":[154],"generate":[157],"action-aware,":[158],"text-enriched":[159],"multimodal":[160],"embeddings.":[161],"Extensive":[162],"experiments":[163],"three":[165],"benchmark":[166],"datasets":[167],"validate":[168],"superior":[170],"performance":[171],"our":[173],"proposed":[174],"approach.":[175]},"counts_by_year":[],"updated_date":"2026-03-09T07:00:12.390032","created_date":"2025-12-09T00:00:00"}