{"id":"https://openalex.org/W4396653437","doi":"https://doi.org/10.1145/3664647.3681034","title":"SATO: Stable Text-to-Motion Framework","display_name":"SATO: Stable Text-to-Motion Framework","publication_year":2024,"publication_date":"2024-10-26","ids":{"openalex":"https://openalex.org/W4396653437","doi":"https://doi.org/10.1145/3664647.3681034"},"language":"en","primary_location":{"id":"doi:10.1145/3664647.3681034","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3664647.3681034","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://hdl.handle.net/1885/733755807","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5003217078","display_name":"Wenshuo Chen","orcid":"https://orcid.org/0009-0001-1062-4509"},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Wenshuo chen","raw_affiliation_strings":["Shandong University, Qingdao, China"],"affiliations":[{"raw_affiliation_string":"Shandong University, Qingdao, China","institution_ids":["https://openalex.org/I154099455"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5096328925","display_name":"Hongru Xiao","orcid":null},"institutions":[{"id":"https://openalex.org/I116953780","display_name":"Tongji University","ror":"https://ror.org/03rc6as71","country_code":"CN","type":"education","lineage":["https://openalex.org/I116953780"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongru Xiao","raw_affiliation_strings":["Tongji University, Shanghai, China"],"affiliations":[{"raw_affiliation_string":"Tongji University, Shanghai, China","institution_ids":["https://openalex.org/I116953780"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5096149162","display_name":"Erhang Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Erhang Zhang","raw_affiliation_strings":["Shandong University, Qingdao, China"],"affiliations":[{"raw_affiliation_string":"Shandong University, Qingdao, China","institution_ids":["https://openalex.org/I154099455"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100929414","display_name":"Lijie Hu","orcid":null},"institutions":[{"id":"https://openalex.org/I71920554","display_name":"King Abdullah University of Science and Technology","ror":"https://ror.org/01q3tbs38","country_code":"SA","type":"education","lineage":["https://openalex.org/I71920554"]}],"countries":["SA"],"is_corresponding":false,"raw_author_name":"Lijie Hu","raw_affiliation_strings":["King Abdullah University of Science and Technology, Jeddah, Saudi Arabia"],"affiliations":[{"raw_affiliation_string":"King Abdullah University of Science and Technology, Jeddah, Saudi Arabia","institution_ids":["https://openalex.org/I71920554"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076665531","display_name":"Lei Wang","orcid":"https://orcid.org/0000-0002-2609-9139"},"institutions":[{"id":"https://openalex.org/I118347636","display_name":"Australian National University","ror":"https://ror.org/019wvm592","country_code":"AU","type":"education","lineage":["https://openalex.org/I118347636"]},{"id":"https://openalex.org/I1292875679","display_name":"Commonwealth Scientific and Industrial Research Organisation","ror":"https://ror.org/03qn8fb07","country_code":"AU","type":"government","lineage":["https://openalex.org/I1292875679","https://openalex.org/I2801453606","https://openalex.org/I4387156119"]},{"id":"https://openalex.org/I42894916","display_name":"Data61","ror":"https://ror.org/03q397159","country_code":"AU","type":"other","lineage":["https://openalex.org/I1292875679","https://openalex.org/I2801453606","https://openalex.org/I42894916","https://openalex.org/I4387156119"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Lei Wang","raw_affiliation_strings":["Australian National University & Data61/CSIRO, Canberra, Australia"],"affiliations":[{"raw_affiliation_string":"Australian National University & Data61/CSIRO, Canberra, Australia","institution_ids":["https://openalex.org/I1292875679","https://openalex.org/I42894916","https://openalex.org/I118347636"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100705476","display_name":"Mengyuan Liu","orcid":"https://orcid.org/0009-0008-7792-9092"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mengyuan Liu","raw_affiliation_strings":["State Key Laboratory of General Artificial Intelligence, Peking University, Shenzhen Graduate School, Shenzhen, China"],"affiliations":[{"raw_affiliation_string":"State Key Laboratory of General Artificial Intelligence, Peking University, Shenzhen Graduate School, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100418433","display_name":"Chen Chen","orcid":"https://orcid.org/0000-0002-3525-9755"},"institutions":[{"id":"https://openalex.org/I106165777","display_name":"University of Central Florida","ror":"https://ror.org/036nfer12","country_code":"US","type":"education","lineage":["https://openalex.org/I106165777"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chen Chen","raw_affiliation_strings":["Center for Research in Computer Vision, University of Central Florida, Orlando, USA"],"affiliations":[{"raw_affiliation_string":"Center for Research in Computer Vision, University of Central Florida, Orlando, USA","institution_ids":["https://openalex.org/I106165777"]}]}],"institutions":[],"countries_distinct_count":4,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5003217078"],"corresponding_institution_ids":["https://openalex.org/I154099455"],"apc_list":null,"apc_paid":null,"fwci":4.8652,"has_fulltext":true,"cited_by_count":14,"citation_normalized_percentile":{"value":0.95527114,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"6989","last_page":"6997"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12031","display_name":"Speech and dialogue systems","score":0.9318000078201294,"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/T12031","display_name":"Speech and dialogue systems","score":0.9318000078201294,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.9205999970436096,"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/motion","display_name":"Motion (physics)","score":0.5528568029403687},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.4516125023365021},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.2675471901893616}],"concepts":[{"id":"https://openalex.org/C104114177","wikidata":"https://www.wikidata.org/wiki/Q79782","display_name":"Motion (physics)","level":2,"score":0.5528568029403687},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4516125023365021},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2675471901893616}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3664647.3681034","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3664647.3681034","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","raw_type":"proceedings-article"},{"id":"pmh:oai:openresearch-repository.anu.edu.au:1885/733755807","is_oa":true,"landing_page_url":"https://hdl.handle.net/1885/733755807","pdf_url":null,"source":{"id":"https://openalex.org/S4306402539","display_name":"ANU Open Research (Australian National University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I118347636","host_organization_name":"Australian National University","host_organization_lineage":["https://openalex.org/I118347636"],"host_organization_lineage_names":[],"type":"repository"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"Conference paper"},{"id":"pmh:oai:arXiv.org:2405.01461","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2405.01461","pdf_url":"https://arxiv.org/pdf/2405.01461","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:openresearch-repository.anu.edu.au:1885/733755807","is_oa":true,"landing_page_url":"https://hdl.handle.net/1885/733755807","pdf_url":null,"source":{"id":"https://openalex.org/S4306402539","display_name":"ANU Open Research (Australian National University)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I118347636","host_organization_name":"Australian National University","host_organization_lineage":["https://openalex.org/I118347636"],"host_organization_lineage_names":[],"type":"repository"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"Conference paper"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":15,"referenced_works":["https://openalex.org/W2474702929","https://openalex.org/W2963083752","https://openalex.org/W3088181395","https://openalex.org/W3144253442","https://openalex.org/W3153832461","https://openalex.org/W4303448003","https://openalex.org/W4312635677","https://openalex.org/W4312936899","https://openalex.org/W4313145975","https://openalex.org/W4386076288","https://openalex.org/W4390872247","https://openalex.org/W4390874627","https://openalex.org/W4391305822","https://openalex.org/W4392909058","https://openalex.org/W4402727293"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W2358668433","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W2382290278","https://openalex.org/W4395014643","https://openalex.org/W4391913857","https://openalex.org/W2350741829"],"abstract_inverted_index":{"Is":[0],"the":[1,24,45,80,91,97,102,120,156,164,167],"Text":[2,10],"to":[3,11,78,132],"Motion":[4,12],"model":[5,94],"robust?":[6],"Recent":[7],"advancements":[8],"in":[9,55],"models":[13,204],"primarily":[14],"stem":[15],"from":[16],"more":[17,188],"accurate":[18],"predictions":[19,49],"of":[20,83,93,101,127,158,166],"specific":[21],"actions.":[22],"However,":[23],"text":[25,69,103],"modality":[26],"typically":[27],"relies":[28],"solely":[29],"on":[30,178],"pre-trained":[31],"Contrastive":[32],"Language-Image":[33],"Pretraining":[34],"(CLIP)":[35],"models.":[36],"Our":[37],"research":[38],"has":[39],"uncovered":[40],"a":[41,87,109,139,148,171],"significant":[42],"issue":[43],"with":[44,64],"text-tomotion":[46],"model:":[47],"its":[48,198],"often":[50],"exhibit":[51],"inconsistent":[52],"outputs,":[53],"resulting":[54],"vastly":[56],"different":[57],"or":[58,67],"even":[59],"incorrect":[60],"poses":[61],"when":[62],"presented":[63],"semantically":[65],"similar":[66],"identical":[68],"inputs.":[70],"In":[71],"this":[72,84,115],"paper,":[73],"we":[74,107,118,169],"undertake":[75],"an":[76,152],"analysis":[77],"elucidate":[79],"underlying":[81],"causes":[82],"instability,":[85],"establishing":[86],"clear":[88],"link":[89],"between":[90,141],"unpredictability":[92],"outputs":[95],"and":[96,137,143,160,180,192,203],"erratic":[98],"attention":[99,159],"patterns":[100],"encoder":[104],"module.":[105],"Consequently,":[106],"introduce":[108],"formal":[110],"framework":[111],"aimed":[112],"at":[113,207],"addressing":[114],"issue,":[116],"which":[117],"term":[119],"Stable":[121],"Text-to-Motion":[122],"Framework":[123],"(SATO).":[124],"SATO":[125,153,185],"consists":[126],"three":[128],"modules,":[129],"each":[130],"dedicated":[131],"stable":[133,135,189],"attention,":[134],"prediction,":[136],"maintaining":[138],"balance":[140],"accuracy":[142,200],"robustness":[144],"trade-off.":[145],"We":[146],"present":[147],"methodology":[149],"for":[150],"constructing":[151],"that":[154,184],"satisfies":[155],"stability":[157,165],"prediction.":[161],"To":[162],"verify":[163],"model,":[168],"introduced":[170],"new":[172],"textual":[173],"synonym":[174],"perturbation":[175],"dataset":[176],"based":[177],"HumanML3D":[179],"KIT-ML.":[181],"Results":[182],"show":[183],"is":[186],"significantly":[187],"against":[190],"synonyms":[191],"other":[193],"slight":[194],"perturbations":[195],"while":[196],"keeping":[197],"high":[199],"performance.":[201],"Codes":[202],"are":[205],"released":[206]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":2}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}