{"id":"https://openalex.org/W4392909833","doi":"https://doi.org/10.1109/icassp48485.2024.10447133","title":"ControlCap: Controllable Captioning via No-Fuss Lexicon","display_name":"ControlCap: Controllable Captioning via No-Fuss Lexicon","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392909833","doi":"https://doi.org/10.1109/icassp48485.2024.10447133"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10447133","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp48485.2024.10447133","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":true,"oa_status":"green","oa_url":"https://figshare.com/articles/conference_contribution/ControlCap_Controllable_Captioning_via_No-Fuss_Lexicon/26361856","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5076958401","display_name":"Qiujie Xie","orcid":"https://orcid.org/0000-0003-0396-9063"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Qiujie Xie","raw_affiliation_strings":["Fudan University,School of Computer Science, Shanghai Key Lab of Intelligent Information Processing,China","School of Computer Science, Shanghai Key Lab of Intelligent Information Processing, Fudan University, China"],"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science, Shanghai Key Lab of Intelligent Information Processing,China","institution_ids":["https://openalex.org/I24943067"]},{"raw_affiliation_string":"School of Computer Science, Shanghai Key Lab of Intelligent Information Processing, Fudan University, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101311538","display_name":"Qiming Feng","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qiming Feng","raw_affiliation_strings":["Fudan University,School of Computer Science, Shanghai Key Lab of Intelligent Information Processing,China","School of Computer Science, Shanghai Key Lab of Intelligent Information Processing, Fudan University, China"],"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science, Shanghai Key Lab of Intelligent Information Processing,China","institution_ids":["https://openalex.org/I24943067"]},{"raw_affiliation_string":"School of Computer Science, Shanghai Key Lab of Intelligent Information Processing, Fudan University, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070073283","display_name":"Yuejie Zhang","orcid":"https://orcid.org/0000-0001-7993-7223"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuejie Zhang","raw_affiliation_strings":["Fudan University,School of Computer Science, Shanghai Key Lab of Intelligent Information Processing,China","School of Computer Science, Shanghai Key Lab of Intelligent Information Processing, Fudan University, China"],"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science, Shanghai Key Lab of Intelligent Information Processing,China","institution_ids":["https://openalex.org/I24943067"]},{"raw_affiliation_string":"School of Computer Science, Shanghai Key Lab of Intelligent Information Processing, Fudan University, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101860531","display_name":"Rui Feng","orcid":"https://orcid.org/0000-0001-6648-953X"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rui Feng","raw_affiliation_strings":["Fudan University,School of Computer Science, Shanghai Key Lab of Intelligent Information Processing,China","School of Computer Science, Shanghai Key Lab of Intelligent Information Processing, Fudan University, China"],"affiliations":[{"raw_affiliation_string":"Fudan University,School of Computer Science, Shanghai Key Lab of Intelligent Information Processing,China","institution_ids":["https://openalex.org/I24943067"]},{"raw_affiliation_string":"School of Computer Science, Shanghai Key Lab of Intelligent Information Processing, Fudan University, China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100375722","display_name":"Tao Zhang","orcid":"https://orcid.org/0000-0001-7561-0143"},"institutions":[{"id":"https://openalex.org/I181679659","display_name":"Shanghai University of Finance and Economics","ror":"https://ror.org/00wtvfq62","country_code":"CN","type":"education","lineage":["https://openalex.org/I181679659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tao Zhang","raw_affiliation_strings":["Shanghai University of Finance and Economics,School of Information Management and Engineering,China","School of Information Management and Engineering, Shanghai University of Finance and Economics, China"],"affiliations":[{"raw_affiliation_string":"Shanghai University of Finance and Economics,School of Information Management and Engineering,China","institution_ids":["https://openalex.org/I181679659"]},{"raw_affiliation_string":"School of Information Management and Engineering, Shanghai University of Finance and Economics, China","institution_ids":["https://openalex.org/I181679659"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5008929449","display_name":"Shang Gao","orcid":"https://orcid.org/0000-0002-2947-7780"},"institutions":[{"id":"https://openalex.org/I149704539","display_name":"Deakin University","ror":"https://ror.org/02czsnj07","country_code":"AU","type":"education","lineage":["https://openalex.org/I149704539"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Shang Gao","raw_affiliation_strings":["Deakin University,School of Information Technology,Australia","School of Information Technology, Deakin University, Australia"],"affiliations":[{"raw_affiliation_string":"Deakin University,School of Information Technology,Australia","institution_ids":["https://openalex.org/I149704539"]},{"raw_affiliation_string":"School of Information Technology, Deakin University, Australia","institution_ids":["https://openalex.org/I149704539"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5076958401"],"corresponding_institution_ids":["https://openalex.org/I24943067"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.02651891,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"8326","last_page":"8330"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9950000047683716,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.9950000047683716,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9509999752044678,"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/T12031","display_name":"Speech and dialogue systems","score":0.9178000092506409,"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/closed-captioning","display_name":"Closed captioning","score":0.8879623413085938},{"id":"https://openalex.org/keywords/lexicon","display_name":"Lexicon","score":0.8012000322341919},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7685204744338989},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4696060121059418},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.458418071269989},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.0416543185710907}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.8879623413085938},{"id":"https://openalex.org/C2778121359","wikidata":"https://www.wikidata.org/wiki/Q8096","display_name":"Lexicon","level":2,"score":0.8012000322341919},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7685204744338989},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4696060121059418},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.458418071269989},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.0416543185710907}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1109/icassp48485.2024.10447133","is_oa":false,"landing_page_url":"http://dx.doi.org/10.1109/icassp48485.2024.10447133","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"},{"id":"pmh:oai:figshare.com:article/26361856","is_oa":true,"landing_page_url":"https://figshare.com/articles/conference_contribution/ControlCap_Controllable_Captioning_via_No-Fuss_Lexicon/26361856","pdf_url":null,"source":{"id":"https://openalex.org/S4377196282","display_name":"Figshare","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4210132348","host_organization_name":"Figshare (United Kingdom)","host_organization_lineage":["https://openalex.org/I4210132348"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Text"}],"best_oa_location":{"id":"pmh:oai:figshare.com:article/26361856","is_oa":true,"landing_page_url":"https://figshare.com/articles/conference_contribution/ControlCap_Controllable_Captioning_via_No-Fuss_Lexicon/26361856","pdf_url":null,"source":{"id":"https://openalex.org/S4377196282","display_name":"Figshare","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4210132348","host_organization_name":"Figshare (United Kingdom)","host_organization_lineage":["https://openalex.org/I4210132348"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Text"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1663965623","display_name":null,"funder_award_id":"2023110139","funder_id":"https://openalex.org/F4320335787","funder_display_name":"Fundamental Research Funds for the Central Universities"},{"id":"https://openalex.org/G2087396116","display_name":null,"funder_award_id":"China","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G2152572007","display_name":null,"funder_award_id":"6197605","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G2372857774","display_name":null,"funder_award_id":"61976057","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G2376276132","display_name":null,"funder_award_id":"China","funder_id":"https://openalex.org/F4320335787","funder_display_name":"Fundamental Research Funds for the Central Universities"},{"id":"https://openalex.org/G3317480652","display_name":null,"funder_award_id":"Science","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G391238517","display_name":null,"funder_award_id":", and","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G4629061667","display_name":null,"funder_award_id":"21511101000","funder_id":"https://openalex.org/F4320321885","funder_display_name":"Science and Technology Commission of Shanghai Municipality"},{"id":"https://openalex.org/G5464079932","display_name":null,"funder_award_id":"62172101","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G5939423041","display_name":null,"funder_award_id":"Technology","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G5994120800","display_name":null,"funder_award_id":"Natural","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320321885","display_name":"Science and Technology Commission of Shanghai Municipality","ror":"https://ror.org/03kt66j61"},{"id":"https://openalex.org/F4320335787","display_name":"Fundamental Research Funds for the Central Universities","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":28,"referenced_works":["https://openalex.org/W68733909","https://openalex.org/W1956340063","https://openalex.org/W2604178507","https://openalex.org/W2625940279","https://openalex.org/W2901988662","https://openalex.org/W2952132648","https://openalex.org/W2963667126","https://openalex.org/W2963916161","https://openalex.org/W2984862483","https://openalex.org/W2997195635","https://openalex.org/W3034655362","https://openalex.org/W3034984754","https://openalex.org/W3093026747","https://openalex.org/W3099215402","https://openalex.org/W4224275552","https://openalex.org/W4283793362","https://openalex.org/W4312289196","https://openalex.org/W4318718936","https://openalex.org/W4382240304","https://openalex.org/W4382459038","https://openalex.org/W4385574358","https://openalex.org/W4386075661","https://openalex.org/W6771522293","https://openalex.org/W6771915120","https://openalex.org/W6777549454","https://openalex.org/W6809844270","https://openalex.org/W6849177959","https://openalex.org/W6898505805"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W4210416330","https://openalex.org/W2775506363","https://openalex.org/W3088136942","https://openalex.org/W4290852288","https://openalex.org/W2949362007","https://openalex.org/W1831473261","https://openalex.org/W4293870971","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Controllable":[0],"captioning":[1,92],"has":[2,12],"received":[3],"much":[4],"attention":[5],"in":[6],"recent":[7],"years.":[8],"Although":[9],"substantial":[10],"progress":[11],"been":[13,87],"made,":[14],"existing":[15],"methods":[16],"still":[17],"face":[18],"challenges":[19],"such":[20],"as":[21,50],"high":[22],"training":[23],"costs,":[24],"intricate":[25],"control":[26,30,51,97],"signals":[27,76],"and":[28,40,53,57,72,84,109],"limited":[29],"capabilities.":[31],"To":[32],"address":[33],"these":[34],"issues,":[35],"we":[36],"propose":[37],"a":[38,47],"straightforward":[39],"unified":[41],"framework":[42],"called":[43],"ControlCap.":[44],"It":[45],"uses":[46],"no-fuss":[48],"lexicon":[49],"signal":[52],"controls":[54],"the":[55,69,96,113],"style":[56],"content":[58,115],"of":[59,99],"visual":[60],"descriptions":[61],"through":[62],"Soft":[63],"Guidance":[64],"(a":[65],"global":[66],"guide":[67],"to":[68],"caption":[70],"distribution)":[71],"Hard":[73],"Force":[74],"(integrating":[75],"without":[77],"additional":[78],"training).":[79],"Extensive":[80],"experiments,":[81],"both":[82],"quantitative":[83],"qualitative,":[85],"have":[86],"conducted":[88],"on":[89],"three":[90],"benchmark":[91],"tasks.":[93],"Results":[94],"demonstrate":[95],"ability":[98],"ControlCap:":[100],"it":[101],"can":[102],"produce":[103],"controlled":[104],"captions":[105],"that":[106],"are":[107],"coherent":[108],"diverse":[110],"while":[111],"keeping":[112],"core":[114],"intact.":[116]},"counts_by_year":[],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2025-10-10T00:00:00"}