{"id":"https://openalex.org/W7164838529","doi":"https://doi.org/10.1145/3805622.3810655","title":"SCAN: Self-Calibrated Textual Anchoring for Dual-Granularity Video Screening in Text-Video Retrieval","display_name":"SCAN: Self-Calibrated Textual Anchoring for Dual-Granularity Video Screening in Text-Video Retrieval","publication_year":2026,"publication_date":"2026-06-15","ids":{"openalex":"https://openalex.org/W7164838529","doi":"https://doi.org/10.1145/3805622.3810655"},"language":null,"primary_location":{"id":"doi:10.1145/3805622.3810655","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805622.3810655","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 International Conference on Multimedia Retrieval","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3805622.3810655","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134182019","display_name":"Dixin Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I139024713","display_name":"Guangdong University of Technology","ror":"https://ror.org/04azbjn80","country_code":"CN","type":"education","lineage":["https://openalex.org/I139024713"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dixin Chen","raw_affiliation_strings":["Guangdong University of Technology, Guangzhou, China"],"raw_orcid":"https://orcid.org/0009-0005-2680-7034","affiliations":[{"raw_affiliation_string":"Guangdong University of Technology, Guangzhou, China","institution_ids":["https://openalex.org/I139024713"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015849144","display_name":"Baoyao Yang","orcid":"https://orcid.org/0000-0001-9092-3164"},"institutions":[{"id":"https://openalex.org/I139024713","display_name":"Guangdong University of Technology","ror":"https://ror.org/04azbjn80","country_code":"CN","type":"education","lineage":["https://openalex.org/I139024713"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Baoyao Yang","raw_affiliation_strings":["Guangdong University of Technology, Guangzhou, China"],"raw_orcid":"https://orcid.org/0000-0001-9092-3164","affiliations":[{"raw_affiliation_string":"Guangdong University of Technology, Guangzhou, China","institution_ids":["https://openalex.org/I139024713"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134147102","display_name":"Haifeng Lin","orcid":null},"institutions":[{"id":"https://openalex.org/I139024713","display_name":"Guangdong University of Technology","ror":"https://ror.org/04azbjn80","country_code":"CN","type":"education","lineage":["https://openalex.org/I139024713"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haifeng Lin","raw_affiliation_strings":["Guangdong University of Technology, Guangzhou, China"],"raw_orcid":"https://orcid.org/0009-0000-7810-9776","affiliations":[{"raw_affiliation_string":"Guangdong University of Technology, Guangzhou, China","institution_ids":["https://openalex.org/I139024713"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138629635","display_name":"Canrong Du","orcid":"https://orcid.org/0009-0001-1826-8510"},"institutions":[{"id":"https://openalex.org/I139024713","display_name":"Guangdong University of Technology","ror":"https://ror.org/04azbjn80","country_code":"CN","type":"education","lineage":["https://openalex.org/I139024713"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Canrong Du","raw_affiliation_strings":["Guangdong University of Technology, Guangzhou, China"],"raw_orcid":"https://orcid.org/0009-0001-1826-8510","affiliations":[{"raw_affiliation_string":"Guangdong University of Technology, Guangzhou, China","institution_ids":["https://openalex.org/I139024713"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5080718178","display_name":"Wenbin Yao","orcid":"https://orcid.org/0000-0002-2830-102X"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wenbin Yao","raw_affiliation_strings":["WeChat, Tencent, Guangzhou, China"],"raw_orcid":"https://orcid.org/0009-0009-5242-6197","affiliations":[{"raw_affiliation_string":"WeChat, Tencent, Guangzhou, China","institution_ids":["https://openalex.org/I2250653659"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"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":0.93767379,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"366","last_page":"375"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9782999753952026,"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":0.9782999753952026,"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/T11439","display_name":"Video Analysis and Summarization","score":0.007000000216066837,"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":0.0031999999191612005,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.6434999704360962},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.5975000262260437},{"id":"https://openalex.org/keywords/video-retrieval","display_name":"Video retrieval","score":0.5748999714851379},{"id":"https://openalex.org/keywords/bridging","display_name":"Bridging (networking)","score":0.5626999735832214},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.4431000053882599},{"id":"https://openalex.org/keywords/anchoring","display_name":"Anchoring","score":0.4278999865055084},{"id":"https://openalex.org/keywords/semantic-gap","display_name":"Semantic gap","score":0.42170000076293945},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.39250001311302185}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8604000210762024},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.6434999704360962},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.5975000262260437},{"id":"https://openalex.org/C2983174267","wikidata":"https://www.wikidata.org/wiki/Q3775098","display_name":"Video retrieval","level":2,"score":0.5748999714851379},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.5637999773025513},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.5626999735832214},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5099999904632568},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4431000053882599},{"id":"https://openalex.org/C18483071","wikidata":"https://www.wikidata.org/wiki/Q168432","display_name":"Anchoring","level":2,"score":0.4278999865055084},{"id":"https://openalex.org/C86034646","wikidata":"https://www.wikidata.org/wiki/Q474311","display_name":"Semantic gap","level":4,"score":0.42170000076293945},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41200000047683716},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.39250001311302185},{"id":"https://openalex.org/C2781122975","wikidata":"https://www.wikidata.org/wiki/Q16928266","display_name":"Semantic feature","level":2,"score":0.3790999948978424},{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.3666999936103821},{"id":"https://openalex.org/C202474056","wikidata":"https://www.wikidata.org/wiki/Q1931635","display_name":"Video tracking","level":3,"score":0.35280001163482666},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.34139999747276306},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3138999938964844},{"id":"https://openalex.org/C2779525943","wikidata":"https://www.wikidata.org/wiki/Q1187300","display_name":"Grammaticality","level":3,"score":0.30160000920295715},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.289000004529953},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.27459999918937683},{"id":"https://openalex.org/C2778493491","wikidata":"https://www.wikidata.org/wiki/Q7449072","display_name":"Semantic matching","level":3,"score":0.2637999951839447},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.26190000772476196},{"id":"https://openalex.org/C49774154","wikidata":"https://www.wikidata.org/wiki/Q131765","display_name":"Multimedia","level":1,"score":0.2533999979496002},{"id":"https://openalex.org/C65483669","wikidata":"https://www.wikidata.org/wiki/Q3536669","display_name":"Video processing","level":2,"score":0.2517000138759613}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3805622.3810655","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805622.3810655","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 International Conference on Multimedia Retrieval","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3805622.3810655","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805622.3810655","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 International Conference on Multimedia Retrieval","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.7092630863189697,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":41,"referenced_works":["https://openalex.org/W2425121537","https://openalex.org/W2902673021","https://openalex.org/W2963017553","https://openalex.org/W2963753226","https://openalex.org/W3034730770","https://openalex.org/W3172655693","https://openalex.org/W3174873881","https://openalex.org/W3204588463","https://openalex.org/W4281560470","https://openalex.org/W4285606530","https://openalex.org/W4312372711","https://openalex.org/W4312884142","https://openalex.org/W4312954271","https://openalex.org/W4312962707","https://openalex.org/W4312999114","https://openalex.org/W4386065852","https://openalex.org/W4386072365","https://openalex.org/W4386076424","https://openalex.org/W4390871725","https://openalex.org/W4390872434","https://openalex.org/W4390873082","https://openalex.org/W4390873165","https://openalex.org/W4390873312","https://openalex.org/W4390889802","https://openalex.org/W4393159202","https://openalex.org/W4393159223","https://openalex.org/W4400525916","https://openalex.org/W4402351150","https://openalex.org/W4402704596","https://openalex.org/W4403791665","https://openalex.org/W4406415185","https://openalex.org/W4409366779","https://openalex.org/W4409368251","https://openalex.org/W4411019516","https://openalex.org/W4411635525","https://openalex.org/W4411635651","https://openalex.org/W4413144376","https://openalex.org/W4413144429","https://openalex.org/W4415709155","https://openalex.org/W4415799009","https://openalex.org/W7160123608"],"related_works":[],"abstract_inverted_index":{"Text-Video":[0],"Retrieval":[1],"(TVR)":[2],"is":[3],"a":[4,20,102,123,162],"foundational":[5],"cross-modal":[6,34],"task":[7],"that":[8,64,107,175],"aims":[9],"to":[10,114,133],"align":[11],"natural":[12],"language":[13],"queries":[14],"with":[15,41],"relevant":[16],"video":[17,27,46,69,104,118,131,159],"content,":[18],"playing":[19],"critical":[21],"role":[22],"in":[23,161,184],"multimedia":[24],"search":[25],"and":[26,76,82,135,145,158,172,189],"understanding.":[28],"Current":[29],"methods":[30],"typically":[31],"address":[32],"the":[33,49,130],"semantic":[35,77,191],"gap":[36],"by":[37],"either":[38],"augmenting":[39],"text":[40,143,157],"generated":[42,59],"captions":[43,60],"or":[44],"enhancing":[45],"representations":[47,70],"at":[48],"feature":[50],"level.":[51],"However,":[52],"these":[53,87],"strategies":[54],"often":[55],"introduce":[56],"new":[57],"challenges:":[58],"may":[61],"contain":[62],"hallucinations":[63],"mislead":[65],"alignment,":[66],"while":[67],"dense":[68],"can":[71],"retain":[72],"substantial":[73],"information":[74,113],"redundancy":[75],"bias,":[78],"collectively":[79],"hindering":[80],"precise":[81],"efficient":[83],"retrieval.":[84,196],"To":[85],"overcome":[86],"limitations,":[88],"we":[89],"propose":[90],"SCAN":[91,176],"(Self-Calibrated":[92],"Textual":[93],"Anchoring":[94],"for":[95,193],"Dual-Granularity":[96],"Video":[97],"Screening).":[98],"Our":[99],"framework":[100],"introduces":[101],"dual-granularity":[103],"screening":[105],"mechanism":[106],"progressively":[108],"filters":[109],"out":[110],"redundant":[111],"visual":[112],"obtain":[115],"semantically":[116],"condensed":[117],"representations.":[119],"Simultaneously,":[120],"it":[121],"employs":[122],"self-calibrated":[124],"textual":[125,138],"anchoring":[126],"module,":[127],"which":[128,155],"leverages":[129],"content":[132],"verify":[134],"reinforce":[136],"reliable":[137],"cues,":[139],"thereby":[140],"strengthening":[141],"core":[142],"semantics":[144,160],"mitigating":[146],"caption":[147],"hallucinations.":[148],"These":[149],"two":[150],"processes":[151],"are":[152],"jointly":[153],"optimized,":[154],"aligns":[156],"synergistic":[163],"manner.":[164],"Extensive":[165],"experiments":[166],"on":[167],"three":[168],"benchmarks":[169],"(MSR-VTT,":[170],"MSVD,":[171],"DiDeMo)":[173],"demonstrate":[174],"consistently":[177],"outperforms":[178],"state-of-the-art":[179],"methods,":[180],"validating":[181],"its":[182],"effectiveness":[183],"reducing":[185],"hallucinations,":[186],"eliminating":[187],"redundancy,":[188],"bridging":[190],"asymmetry":[192],"robust":[194],"text-video":[195]},"counts_by_year":[],"updated_date":"2026-06-16T07:37:23.134862","created_date":"2026-06-16T00:00:00"}