{"id":"https://openalex.org/W4392904725","doi":"https://doi.org/10.1109/icassp48485.2024.10446193","title":"WAVER: Writing-Style Agnostic Text-Video Retrieval Via Distilling Vision-Language Models Through Open-Vocabulary Knowledge","display_name":"WAVER: Writing-Style Agnostic Text-Video Retrieval Via Distilling Vision-Language Models Through Open-Vocabulary Knowledge","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392904725","doi":"https://doi.org/10.1109/icassp48485.2024.10446193"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10446193","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10446193","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","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/A5078515449","display_name":"Huy Quoc Le","orcid":"https://orcid.org/0000-0003-4862-6048"},"institutions":[{"id":"https://openalex.org/I883909435","display_name":"International University","ror":"https://ror.org/02knr1992","country_code":"KH","type":"education","lineage":["https://openalex.org/I883909435"]},{"id":"https://openalex.org/I109689652","display_name":"FPT University","ror":"https://ror.org/03esj4g97","country_code":"VN","type":"education","lineage":["https://openalex.org/I109689652"]}],"countries":["KH","VN"],"is_corresponding":true,"raw_author_name":"Huy Le","raw_affiliation_strings":["FPT Software AI Center","International University, VNU-HCM"],"affiliations":[{"raw_affiliation_string":"FPT Software AI Center","institution_ids":["https://openalex.org/I109689652"]},{"raw_affiliation_string":"International University, VNU-HCM","institution_ids":["https://openalex.org/I883909435"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016799625","display_name":"Tung Kieu","orcid":"https://orcid.org/0000-0002-7696-1444"},"institutions":[{"id":"https://openalex.org/I157358134","display_name":"RMIT Vietnam","ror":"https://ror.org/004axh929","country_code":"VN","type":"education","lineage":["https://openalex.org/I157358134","https://openalex.org/I82951845"]}],"countries":["VN"],"is_corresponding":false,"raw_author_name":"Tung Kieu","raw_affiliation_strings":["RMIT University,Vietnam","RMIT University, Vietnam"],"affiliations":[{"raw_affiliation_string":"RMIT University,Vietnam","institution_ids":["https://openalex.org/I157358134"]},{"raw_affiliation_string":"RMIT University, Vietnam","institution_ids":["https://openalex.org/I157358134"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089006585","display_name":"Anh Nguyen","orcid":"https://orcid.org/0000-0002-1449-211X"},"institutions":[{"id":"https://openalex.org/I146655781","display_name":"University of Liverpool","ror":"https://ror.org/04xs57h96","country_code":"GB","type":"education","lineage":["https://openalex.org/I146655781"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Anh Nguyen","raw_affiliation_strings":["University of Liverpool,UK","University of Liverpool, UK"],"affiliations":[{"raw_affiliation_string":"University of Liverpool,UK","institution_ids":["https://openalex.org/I146655781"]},{"raw_affiliation_string":"University of Liverpool, UK","institution_ids":["https://openalex.org/I146655781"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5023725893","display_name":"Ngan Le","orcid":"https://orcid.org/0000-0003-2571-0511"},"institutions":[{"id":"https://openalex.org/I78715868","display_name":"University of Arkansas at Fayetteville","ror":"https://ror.org/05jbt9m15","country_code":"US","type":"education","lineage":["https://openalex.org/I78715868"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ngan Le","raw_affiliation_strings":["University of Arkansas,USA","University of Arkansas, USA"],"affiliations":[{"raw_affiliation_string":"University of Arkansas,USA","institution_ids":["https://openalex.org/I78715868"]},{"raw_affiliation_string":"University of Arkansas, USA","institution_ids":["https://openalex.org/I78715868"]}]}],"institutions":[],"countries_distinct_count":4,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5078515449"],"corresponding_institution_ids":["https://openalex.org/I109689652","https://openalex.org/I883909435"],"apc_list":null,"apc_paid":null,"fwci":0.7895,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.70212501,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"3025","last_page":"3029"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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":1.0,"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/T10028","display_name":"Topic Modeling","score":0.9976000189781189,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9958999752998352,"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/computer-science","display_name":"Computer science","score":0.8399927616119385},{"id":"https://openalex.org/keywords/vocabulary","display_name":"Vocabulary","score":0.727056622505188},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.6045444011688232},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5794066786766052},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.5597911477088928},{"id":"https://openalex.org/keywords/style","display_name":"Style (visual arts)","score":0.5496639609336853},{"id":"https://openalex.org/keywords/writing-style","display_name":"Writing style","score":0.5496323704719543},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5157337784767151},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4357856810092926},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.42419907450675964},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.41722697019577026},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.4083822965621948},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.17031964659690857},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.08332881331443787}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8399927616119385},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.727056622505188},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6045444011688232},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5794066786766052},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.5597911477088928},{"id":"https://openalex.org/C2776445246","wikidata":"https://www.wikidata.org/wiki/Q1792644","display_name":"Style (visual arts)","level":2,"score":0.5496639609336853},{"id":"https://openalex.org/C13622073","wikidata":"https://www.wikidata.org/wiki/Q2243831","display_name":"Writing style","level":2,"score":0.5496323704719543},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5157337784767151},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4357856810092926},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.42419907450675964},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.41722697019577026},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.4083822965621948},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.17031964659690857},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.08332881331443787},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.0},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10446193","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10446193","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.8399999737739563,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":48,"referenced_works":["https://openalex.org/W1821462560","https://openalex.org/W2425121537","https://openalex.org/W2842511635","https://openalex.org/W2963017553","https://openalex.org/W2982242214","https://openalex.org/W2989322838","https://openalex.org/W3035356601","https://openalex.org/W3043840704","https://openalex.org/W3151130473","https://openalex.org/W3168640669","https://openalex.org/W3176799298","https://openalex.org/W3197804339","https://openalex.org/W3204588463","https://openalex.org/W3212304713","https://openalex.org/W4221142658","https://openalex.org/W4221146248","https://openalex.org/W4225414521","https://openalex.org/W4226058394","https://openalex.org/W4285606530","https://openalex.org/W4295312788","https://openalex.org/W4297808394","https://openalex.org/W4297947725","https://openalex.org/W4312299780","https://openalex.org/W4312364543","https://openalex.org/W4312372711","https://openalex.org/W4312691946","https://openalex.org/W4312912313","https://openalex.org/W4312999114","https://openalex.org/W4313136445","https://openalex.org/W4381328595","https://openalex.org/W4386076600","https://openalex.org/W4386076615","https://openalex.org/W6638523607","https://openalex.org/W6684090549","https://openalex.org/W6766582784","https://openalex.org/W6766978945","https://openalex.org/W6784184991","https://openalex.org/W6784333009","https://openalex.org/W6790019176","https://openalex.org/W6791353385","https://openalex.org/W6797148833","https://openalex.org/W6801013943","https://openalex.org/W6803545775","https://openalex.org/W6810042059","https://openalex.org/W6811013733","https://openalex.org/W6839632867","https://openalex.org/W6843148291","https://openalex.org/W6849177959"],"related_works":["https://openalex.org/W4319265185","https://openalex.org/W2357762156","https://openalex.org/W1990462918","https://openalex.org/W2353426914","https://openalex.org/W2352254211","https://openalex.org/W2393938063","https://openalex.org/W2384332983","https://openalex.org/W2910453172","https://openalex.org/W2374347229","https://openalex.org/W2389992592"],"abstract_inverted_index":{"Text-video":[0],"retrieval,":[1,11],"a":[2,61,109,113],"prominent":[3],"sub-field":[4],"within":[5],"the":[6,55,75,88],"domain":[7],"of":[8,77],"multimodal":[9],"information":[10],"has":[12],"witnessed":[13],"remarkable":[14],"growth":[15],"in":[16,82,93,136],"recent":[17],"years.":[18],"However,":[19],"existing":[20],"methods":[21],"assume":[22],"video":[23,83],"scenes":[24],"are":[25],"consistent":[26],"with":[27,35],"unbiased":[28],"descriptions.":[29,84],"These":[30],"limitations":[31],"fail":[32],"to":[33,73,104,112],"align":[34],"real-world":[36],"scenarios":[37],"since":[38],"descriptions":[39],"can":[40,132],"be":[41],"influenced":[42],"by":[43],"annotator":[44],"biases,":[45],"diverse":[46],"writing":[47,80],"styles,":[48],"and":[49,97],"varying":[50],"textual":[51],"perspectives.":[52],"To":[53],"overcome":[54],"aforementioned":[56],"problems,":[57],"we":[58],"introduce":[59],"WAVER,":[60],"cross-domain":[62],"knowledge":[63,71,101,107],"distillation":[64,102],"framework":[65],"via":[66],"vision-language":[67,95],"models":[68,96],"through":[69],"open-vocabulary":[70,89],"designed":[72],"tackle":[74],"challenge":[76],"handling":[78,141],"different":[79],"styles":[81],"WAVER":[85,131],"capitalizes":[86],"on":[87],"properties":[90],"that":[91,130],"lie":[92],"pre-trained":[94],"employs":[98],"an":[99],"implicit":[100],"approach":[103],"transfer":[105],"text-based":[106],"from":[108],"teacher":[110],"model":[111],"vision-based":[114],"student.":[115],"Empirical":[116],"studies":[117],"conducted":[118],"across":[119],"four":[120],"standard":[121],"benchmark":[122],"datasets,":[123],"encompassing":[124],"various":[125],"settings,":[126],"provide":[127],"compelling":[128],"evidence":[129],"achieve":[133],"state-of-the-art":[134],"performance":[135],"text-video":[137],"retrieval":[138],"task":[139],"while":[140],"writing-style":[142],"variations.":[143],"The":[144],"code":[145],"is":[146],"available":[147],"at:":[148],"https://github.com/Fsoft-AIC/WAVER":[149]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}