{"id":"https://openalex.org/W4412945111","doi":"https://doi.org/10.18653/v1/2025.acl-long.807","title":"Task-Specific Information Decomposition for End-to-End Dense Video Captioning","display_name":"Task-Specific Information Decomposition for End-to-End Dense Video Captioning","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412945111","doi":"https://doi.org/10.18653/v1/2025.acl-long.807"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.807","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.807","pdf_url":"https://aclanthology.org/2025.acl-long.807.pdf","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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.807.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5104154574","display_name":"Zhiyue Liu","orcid":"https://orcid.org/0000-0001-7989-7348"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhiyue Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102017129","display_name":"Xinru Zhang","orcid":"https://orcid.org/0009-0002-5390-0566"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xinru Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100675903","display_name":"Jinyuan Liu","orcid":"https://orcid.org/0000-0003-0488-0263"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinyuan Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.15001629,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"16524","last_page":"16536"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9990000128746033,"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":0.9990000128746033,"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":0.9965999722480774,"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/T10531","display_name":"Advanced Vision and Imaging","score":0.9962999820709229,"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/closed-captioning","display_name":"Closed captioning","score":0.7660727500915527},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7277829647064209},{"id":"https://openalex.org/keywords/end-to-end-principle","display_name":"End-to-end principle","score":0.7244752645492554},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6903458833694458},{"id":"https://openalex.org/keywords/decomposition","display_name":"Decomposition","score":0.45583203434944153},{"id":"https://openalex.org/keywords/end-user","display_name":"End user","score":0.4515940845012665},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.32539618015289307},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3149015009403229},{"id":"https://openalex.org/keywords/world-wide-web","display_name":"World Wide Web","score":0.21502798795700073},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.10529708862304688},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.0682026743888855}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.7660727500915527},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7277829647064209},{"id":"https://openalex.org/C74296488","wikidata":"https://www.wikidata.org/wiki/Q2527392","display_name":"End-to-end principle","level":2,"score":0.7244752645492554},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6903458833694458},{"id":"https://openalex.org/C124681953","wikidata":"https://www.wikidata.org/wiki/Q339062","display_name":"Decomposition","level":2,"score":0.45583203434944153},{"id":"https://openalex.org/C91262260","wikidata":"https://www.wikidata.org/wiki/Q528074","display_name":"End user","level":2,"score":0.4515940845012665},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.32539618015289307},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3149015009403229},{"id":"https://openalex.org/C136764020","wikidata":"https://www.wikidata.org/wiki/Q466","display_name":"World Wide Web","level":1,"score":0.21502798795700073},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.10529708862304688},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.0682026743888855},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0},{"id":"https://openalex.org/C18903297","wikidata":"https://www.wikidata.org/wiki/Q7150","display_name":"Ecology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.807","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.807","pdf_url":"https://aclanthology.org/2025.acl-long.807.pdf","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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.807","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.807","pdf_url":"https://aclanthology.org/2025.acl-long.807.pdf","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 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.4099999964237213,"id":"https://metadata.un.org/sdg/13","display_name":"Climate action"}],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320322768","display_name":"Natural Science Foundation of Guangxi Province","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412945111.pdf","grobid_xml":"https://content.openalex.org/works/W4412945111.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4299590256","https://openalex.org/W2151749779","https://openalex.org/W4404782863","https://openalex.org/W3163634122","https://openalex.org/W2110442089","https://openalex.org/W2166381389","https://openalex.org/W4315785295","https://openalex.org/W4393280045","https://openalex.org/W2054736184","https://openalex.org/W3159728998"],"abstract_inverted_index":{"Dense":[0],"video":[1,64,73],"captioning":[2,74,80,115,148],"aims":[3],"to":[4,23,50,108,117,134,142,150],"localize":[5],"events":[6],"within":[7],"input":[8],"videos":[9],"and":[10,58,79,114,147,161],"generate":[11],"concise":[12],"descriptive":[13],"texts":[14],"for":[15,130],"each":[16],"event.Advanced":[17],"endto-end":[18],"methods":[19,129],"require":[20],"both":[21],"tasks":[22],"share":[24],"the":[25,36,47,92,136,139,159],"same":[26],"intermediate":[27],"features":[28,113],"that":[29,76,104,166],"serve":[30],"as":[31,54],"event":[32,55,83],"queries,":[33,84,96],"thereby":[34],"enabling":[35,85],"mutual":[37],"promotion":[38],"of":[39,94,138],"two":[40],"tasks.However,":[41],"relying":[42],"on":[43,63,110,158],"shared":[44],"queries":[45,81,107,116,141],"limits":[46],"model's":[48],"ability":[49],"extract":[51],"taskspecific":[52],"information,":[53],"semantic":[56,101],"perception":[57],"localization":[59,78,106,123,146],"demand":[60],"distinct":[61],"perspectives":[62],"understanding.To":[65],"address":[66],"this,":[67],"we":[68,97],"propose":[69],"a":[70,99,152],"decomposed":[71],"dense":[72],"framework":[75,168],"derives":[77],"from":[82],"task-specific":[86],"representations":[87],"while":[88],"maintaining":[89],"inter-task":[90],"collaboration.Considering":[91],"roles":[93],"different":[95],"design":[98],"contrastive":[100],"optimization":[102],"strategy":[103],"guides":[105],"focus":[109],"event-level":[111],"visual":[112],"align":[118],"with":[119],"textual":[120],"semantics.Besides,":[121],"only":[122],"information":[124],"is":[125],"considered":[126],"in":[127],"existing":[128],"label":[131],"assignment,":[132],"failing":[133],"ensure":[135],"relevance":[137],"selected":[140],"descriptions.We":[143],"jointly":[144],"consider":[145],"losses":[149],"achieve":[151],"semantically":[153],"balanced":[154],"assignment":[155],"process.Extensive":[156],"experiments":[157],"YouCook2":[160],"ActivityNet":[162],"Captions":[163],"datasets":[164],"demonstrate":[165],"our":[167],"achieves":[169],"state-of-the-art":[170],"performance.":[171]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}