{"id":"https://openalex.org/W3088548005","doi":"https://doi.org/10.18653/v1/2020.aacl-main.50","title":"Are Scene Graphs Good Enough to Improve Image Captioning?","display_name":"Are Scene Graphs Good Enough to Improve Image Captioning?","publication_year":2020,"publication_date":"2020-01-01","ids":{"openalex":"https://openalex.org/W3088548005","doi":"https://doi.org/10.18653/v1/2020.aacl-main.50","mag":"3088548005"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2020.aacl-main.50","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2020.aacl-main.50","pdf_url":"https://aclanthology.org/2020.aacl-main.50.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 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing","raw_type":"proceedings-article"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2020.aacl-main.50.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5090907746","display_name":"Victor Milewski","orcid":"https://orcid.org/0000-0003-3854-6386"},"institutions":[{"id":"https://openalex.org/I99464096","display_name":"KU Leuven","ror":"https://ror.org/05f950310","country_code":"BE","type":"education","lineage":["https://openalex.org/I99464096"]}],"countries":["BE"],"is_corresponding":true,"raw_author_name":"Victor Milewski","raw_affiliation_strings":["Katholieke Universiteit Leuven"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Katholieke Universiteit Leuven","institution_ids":["https://openalex.org/I99464096"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075796989","display_name":"Marie\u2010Francine Moens","orcid":"https://orcid.org/0000-0002-3732-9323"},"institutions":[{"id":"https://openalex.org/I99464096","display_name":"KU Leuven","ror":"https://ror.org/05f950310","country_code":"BE","type":"education","lineage":["https://openalex.org/I99464096"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Marie-Francine Moens","raw_affiliation_strings":["Katholieke Universiteit Leuven"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Katholieke Universiteit Leuven","institution_ids":["https://openalex.org/I99464096"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5086777036","display_name":"Iacer Calixto","orcid":null},"institutions":[{"id":"https://openalex.org/I102475099","display_name":"Sup\u00e9lec","ror":"https://ror.org/00n7gwn90","country_code":"FR","type":"education","lineage":["https://openalex.org/I102475099"]},{"id":"https://openalex.org/I15196421","display_name":"University of Applied Sciences and Arts of Southern Switzerland","ror":"https://ror.org/05ep8g269","country_code":"CH","type":"education","lineage":["https://openalex.org/I15196421"]},{"id":"https://openalex.org/I4210102264","display_name":"Shandong University of Political Science and Law","ror":"https://ror.org/01b2j5886","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210102264"]}],"countries":["CH","CN","FR"],"is_corresponding":false,"raw_author_name":"Iacer Calixto","raw_affiliation_strings":["New York University<sup>**</sup>"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"New York University<sup>**</sup>","institution_ids":["https://openalex.org/I15196421","https://openalex.org/I4210102264","https://openalex.org/I102475099"]}]}],"institutions":[],"countries_distinct_count":4,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5090907746"],"corresponding_institution_ids":["https://openalex.org/I99464096"],"apc_list":null,"apc_paid":null,"fwci":0.2942,"has_fulltext":true,"cited_by_count":3,"citation_normalized_percentile":{"value":0.57262098,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"504","last_page":"515"},"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9941999912261963,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9879999756813049,"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":0.9709298014640808},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7851084470748901},{"id":"https://openalex.org/keywords/scene-graph","display_name":"Scene graph","score":0.7462574243545532},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.5572667717933655},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5457916855812073},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5229376554489136},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.5195122957229614},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5105496644973755},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.41015559434890747},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.27620089054107666}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.9709298014640808},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7851084470748901},{"id":"https://openalex.org/C179372163","wikidata":"https://www.wikidata.org/wiki/Q1406181","display_name":"Scene graph","level":3,"score":0.7462574243545532},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.5572667717933655},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5457916855812073},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5229376554489136},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.5195122957229614},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5105496644973755},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.41015559434890747},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.27620089054107666},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C205711294","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Rendering (computer graphics)","level":2,"score":0.0}],"mesh":[],"locations_count":7,"locations":[{"id":"doi:10.18653/v1/2020.aacl-main.50","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2020.aacl-main.50","pdf_url":"https://aclanthology.org/2020.aacl-main.50.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 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2009.12313","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2009.12313","pdf_url":"https://arxiv.org/pdf/2009.12313","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"pmh:oai:dare.uva.nl:publications/0af94bf2-145f-4a37-9aa4-7f13f73d5a0a","is_oa":true,"landing_page_url":"https://handle.uba.uva.nl/personal/pure/en/publications/are-scene-graphs-good-enough-to-improve-image-captioning(0af94bf2-145f-4a37-9aa4-7f13f73d5a0a).html","pdf_url":"https://dare.uva.nl/personal/pure/en/publications/are-scene-graphs-good-enough-to-improve-image-captioning(0af94bf2-145f-4a37-9aa4-7f13f73d5a0a).html","source":{"id":"https://openalex.org/S4306400088","display_name":"UvA-DARE (University of Amsterdam)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I887064364","host_organization_name":"University of Amsterdam","host_organization_lineage":["https://openalex.org/I887064364"],"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":"Milewski, V, Moens, M-F & Calixto, I 2020, Are Scene Graphs Good Enough to Improve Image Captioning? in K-F Wong, K Knight & H Wu (eds), The 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing : AACL-IJCNLP 2020 : proceedings of the conference : December 4-7, 2020. Stroudsburg, PA, pp. 504-515, 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and 10th International Joint Conference on Natural Language Processing, Suzhou, China, 4/12/20. < https://www.aclweb.org/anthology/2020.aacl-main.50/ >","raw_type":"info:eu-repo/semantics/publishedVersion"},{"id":"pmh:oai:lirias2repo.kuleuven.be:123456789/675434","is_oa":true,"landing_page_url":"https://lirias.kuleuven.be/handle/123456789/675434","pdf_url":"https://lirias.kuleuven.be/bitstream/123456789/675434/2/2020.aacl-main.50.pdf","source":{"id":"https://openalex.org/S4306401954","display_name":"Lirias (KU Leuven)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I99464096","host_organization_name":"KU Leuven","host_organization_lineage":["https://openalex.org/I99464096"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Asian Chapter of the Association for Computational Linguistics (AACL), Suzhou, China, 4-7 December 2020","raw_type":"info:eu-repo/semantics/publishedVersion"},{"id":"mag:3088548005","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/2009.12313.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"pmh:uvapub:oai:dare.uva.nl:publications/0af94bf2-145f-4a37-9aa4-7f13f73d5a0a","is_oa":true,"landing_page_url":"https://dare.uva.nl/personal/pure/en/publications/are-scene-graphs-good-enough-to-improve-image-captioning(0af94bf2-145f-4a37-9aa4-7f13f73d5a0a).html","pdf_url":null,"source":{"id":"https://openalex.org/S4306401843","display_name":"Data Archiving and Networked Services (DANS)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1322597698","host_organization_name":"Royal Netherlands Academy of Arts and Sciences","host_organization_lineage":["https://openalex.org/I1322597698"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"The 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing: AACL-IJCNLP 2020 : proceedings of the conference : December 4-7, 2020, 504 - 515","raw_type":"info:eu-repo/semantics/conferencepaper"},{"id":"doi:10.48550/arxiv.2009.12313","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2009.12313","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.18653/v1/2020.aacl-main.50","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2020.aacl-main.50","pdf_url":"https://aclanthology.org/2020.aacl-main.50.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 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G446009141","display_name":"Commonsense and Anticipation enriched Learning of Continuous representations sUpporting Language UnderStanding","funder_award_id":"788506","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G6547846016","display_name":"IMAGINE \u2013 Informing Multi-modal lAnguage Generation wIth world kNowledgE","funder_award_id":"838188","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"},{"id":"https://openalex.org/G8788504996","display_name":null,"funder_award_id":"CA18231","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"}],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"}],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3088548005.pdf"},"referenced_works_count":36,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W1536680647","https://openalex.org/W1861492603","https://openalex.org/W1956340063","https://openalex.org/W2064675550","https://openalex.org/W2077069816","https://openalex.org/W2101105183","https://openalex.org/W2102605133","https://openalex.org/W2154652894","https://openalex.org/W2194775991","https://openalex.org/W2277195237","https://openalex.org/W2506483933","https://openalex.org/W2561715562","https://openalex.org/W2575842049","https://openalex.org/W2579549467","https://openalex.org/W2613718673","https://openalex.org/W2745461083","https://openalex.org/W2795151422","https://openalex.org/W2805516822","https://openalex.org/W2810482788","https://openalex.org/W2886970679","https://openalex.org/W2911710347","https://openalex.org/W2913618459","https://openalex.org/W2963068946","https://openalex.org/W2963084599","https://openalex.org/W2963184176","https://openalex.org/W2963403868","https://openalex.org/W2963435215","https://openalex.org/W2963536419","https://openalex.org/W2963653811","https://openalex.org/W2963858333","https://openalex.org/W2964308564","https://openalex.org/W2981165461","https://openalex.org/W2989377923","https://openalex.org/W2990818246","https://openalex.org/W3025726122"],"related_works":["https://openalex.org/W3118006287","https://openalex.org/W3202620359","https://openalex.org/W3127546394","https://openalex.org/W3184581815","https://openalex.org/W2992478697","https://openalex.org/W3194910448","https://openalex.org/W2891658588","https://openalex.org/W3092867601","https://openalex.org/W2963649796","https://openalex.org/W2950626540","https://openalex.org/W2952342379","https://openalex.org/W1938708284","https://openalex.org/W3204462647","https://openalex.org/W3201591743","https://openalex.org/W2928594414","https://openalex.org/W3168954372","https://openalex.org/W2785567709","https://openalex.org/W2913090763","https://openalex.org/W3208872528","https://openalex.org/W2889720786"],"abstract_inverted_index":{"Victor":[0],"Milewski,":[1],"Marie-Francine":[2],"Moens,":[3],"Iacer":[4],"Calixto.":[5],"Proceedings":[6],"of":[7,11,15],"the":[8,12,16,22],"1st":[9],"Conference":[10,26],"Asia-Pacific":[13],"Chapter":[14],"Association":[17],"for":[18],"Computational":[19],"Linguistics":[20],"and":[21],"10th":[23],"International":[24],"Joint":[25],"on":[27],"Natural":[28],"Language":[29],"Processing.":[30],"2020.":[31]},"counts_by_year":[{"year":2021,"cited_by_count":3}],"updated_date":"2026-05-04T08:30:34.212998","created_date":"2025-10-10T00:00:00"}