{"id":"https://openalex.org/W4413146568","doi":"https://doi.org/10.1109/cvpr52734.2025.01288","title":"SVLTA: Benchmarking Vision-Language Temporal Alignment via Synthetic Video Situation","display_name":"SVLTA: Benchmarking Vision-Language Temporal Alignment via Synthetic Video Situation","publication_year":2025,"publication_date":"2025-06-10","ids":{"openalex":"https://openalex.org/W4413146568","doi":"https://doi.org/10.1109/cvpr52734.2025.01288"},"language":"en","primary_location":{"id":"doi:10.1109/cvpr52734.2025.01288","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52734.2025.01288","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"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/A5044052726","display_name":"Hao Du","orcid":"https://orcid.org/0000-0002-0551-2377"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Du","raw_affiliation_strings":["University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100404779","display_name":"Bo Wu","orcid":"https://orcid.org/0000-0002-2558-8360"},"institutions":[{"id":"https://openalex.org/I1341412227","display_name":"IBM (United States)","ror":"https://ror.org/05hh8d621","country_code":"US","type":"company","lineage":["https://openalex.org/I1341412227"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Bo Wu","raw_affiliation_strings":["MIT-IBM Watson AI Lab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"MIT-IBM Watson AI Lab","institution_ids":["https://openalex.org/I1341412227"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055248712","display_name":"Yan Lu","orcid":"https://orcid.org/0000-0001-6608-003X"},"institutions":[{"id":"https://openalex.org/I177725633","display_name":"Chinese University of Hong Kong","ror":"https://ror.org/00t33hh48","country_code":"HK","type":"education","lineage":["https://openalex.org/I177725633"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yan Lu","raw_affiliation_strings":["The Chinese University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong","institution_ids":["https://openalex.org/I177725633"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5023341829","display_name":"Zhendong Mao","orcid":"https://orcid.org/0000-0001-5739-8126"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhendong Mao","raw_affiliation_strings":["University of Science and Technology of China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China","institution_ids":["https://openalex.org/I126520041"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"13798","last_page":"13809"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9987999796867371,"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":0.9987999796867371,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9908000230789185,"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/T13310","display_name":"Subtitles and Audiovisual Media","score":0.9776999950408936,"subfield":{"id":"https://openalex.org/subfields/1203","display_name":"Language and Linguistics"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.8507105112075806},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7747312188148499},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5956248641014099},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.5359770655632019}],"concepts":[{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.8507105112075806},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7747312188148499},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5956248641014099},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.5359770655632019},{"id":"https://openalex.org/C162853370","wikidata":"https://www.wikidata.org/wiki/Q39809","display_name":"Marketing","level":1,"score":0.0},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/cvpr52734.2025.01288","is_oa":false,"landing_page_url":"https://doi.org/10.1109/cvpr52734.2025.01288","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/13","display_name":"Climate action","score":0.41999998688697815}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":56,"referenced_works":["https://openalex.org/W2111078031","https://openalex.org/W2397830550","https://openalex.org/W2799002257","https://openalex.org/W2890502146","https://openalex.org/W2899004870","https://openalex.org/W2948589675","https://openalex.org/W2962778061","https://openalex.org/W2963017553","https://openalex.org/W2963393391","https://openalex.org/W2963916161","https://openalex.org/W2964089981","https://openalex.org/W2970898753","https://openalex.org/W2991727786","https://openalex.org/W2996908057","https://openalex.org/W2997429269","https://openalex.org/W3034679267","https://openalex.org/W3034743747","https://openalex.org/W3035339529","https://openalex.org/W3035356601","https://openalex.org/W3035640828","https://openalex.org/W3096121526","https://openalex.org/W3104862079","https://openalex.org/W3106723659","https://openalex.org/W3119150854","https://openalex.org/W3152619510","https://openalex.org/W3174490084","https://openalex.org/W3175817778","https://openalex.org/W3187433838","https://openalex.org/W3205786327","https://openalex.org/W3216763528","https://openalex.org/W3216765867","https://openalex.org/W4304086137","https://openalex.org/W4312245888","https://openalex.org/W4312271977","https://openalex.org/W4312341083","https://openalex.org/W4312467626","https://openalex.org/W4313071966","https://openalex.org/W4319301030","https://openalex.org/W4382462100","https://openalex.org/W4386065512","https://openalex.org/W4386075757","https://openalex.org/W4386083094","https://openalex.org/W4390872505","https://openalex.org/W4390873341","https://openalex.org/W4402660113","https://openalex.org/W4402670320","https://openalex.org/W4402671548","https://openalex.org/W4402671847","https://openalex.org/W4402716035","https://openalex.org/W4402716477","https://openalex.org/W4402727142","https://openalex.org/W4402727272","https://openalex.org/W4402754238","https://openalex.org/W4403081466","https://openalex.org/W4404784276","https://openalex.org/W4411244918"],"related_works":["https://openalex.org/W2772917594","https://openalex.org/W2036807459","https://openalex.org/W2058170566","https://openalex.org/W2755342338","https://openalex.org/W2166024367","https://openalex.org/W3116076068","https://openalex.org/W2229312674","https://openalex.org/W2951359407","https://openalex.org/W2079911747","https://openalex.org/W1969923398"],"abstract_inverted_index":{"Vision-language":[0],"temporal":[1,30,60,154,161],"alignment":[2,57,162],"is":[3,46],"a":[4,59,75,80,97,113,121],"crucial":[5],"capability":[6],"for":[7,142],"human":[8],"dynamic":[9],"recognition":[10],"and":[11,34,41,49,91,115,131,138,160],"cognition":[12],"in":[13,74,153],"real-world":[14],"scenarios.":[15],"While":[16],"existing":[17,89,94],"research":[18],"focuses":[19],"on":[20,64],"capturing":[21],"vision-language":[22],"relevance,":[23],"it":[24],"faces":[25],"limitations":[26],"due":[27],"to":[28,47,55,67],"biased":[29],"distributions,":[31],"imprecise":[32],"annotations,":[33],"insufficient":[35],"compositionally.":[36],"To":[37,100],"achieve":[38,56],"fair":[39],"evaluation":[40],"comprehensive":[42],"exploration,":[43],"our":[44],"objective":[45],"investigate":[48],"evaluate":[50],"the":[51,85,93,106,151],"ability":[52],"of":[53,88],"models":[54],"from":[58,96],"perspective,":[61],"specifically":[62],"focusing":[63],"their":[65],"capacity":[66],"synchronize":[68],"visual":[69],"scenarios":[70],"with":[71],"linguistic":[72],"context":[73],"temporally":[76],"coherent":[77],"manner.":[78],"As":[79],"preliminary":[81],"step,":[82],"we":[83,103],"present":[84],"statistical":[86],"analysis":[87],"benchmarks":[90],"reveal":[92,147],"challenges":[95],"decomposed":[98],"perspective.":[99],"this":[101],"end,":[102],"introduce":[104],"SVLTA,":[105],"Synthetic":[107],"Vision-Language":[108],"Temporal":[109],"Alignment":[110],"derived":[111],"via":[112],"well-designed":[114],"feasible":[116],"control":[117],"generation":[118],"method":[119],"within":[120],"simulation":[122],"environment.":[123],"The":[124],"approach":[125],"considers":[126],"commonsense":[127],"knowledge,":[128],"manipulable":[129],"action,":[130],"constrained":[132],"filtering,":[133],"which":[134],"generates":[135],"reasonable,":[136],"diverse,":[137],"balanced":[139],"data":[140],"distributions":[141],"diagnostic":[143,148],"evaluations.":[144],"Our":[145],"experiments":[146],"insights":[149],"through":[150],"evaluations":[152],"question":[155],"answering,":[156],"distributional":[157],"shift":[158],"sensitiveness,":[159],"adaptation.":[163]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}