{"id":"https://openalex.org/W4412944495","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1367","title":"EXPERT: An Explainable Image Captioning Evaluation Metric with Structured Explanations","display_name":"EXPERT: An Explainable Image Captioning Evaluation Metric with Structured Explanations","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412944495","doi":"https://doi.org/10.18653/v1/2025.findings-acl.1367"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.1367","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1367","pdf_url":"https://aclanthology.org/2025.findings-acl.1367.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.1367.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101940819","display_name":"Hyunjong Kim","orcid":"https://orcid.org/0000-0001-6254-663X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hyunjong Kim","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063424739","display_name":"Sangyeop Kim","orcid":"https://orcid.org/0000-0002-7584-1061"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sangyeop Kim","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045815823","display_name":"Jongheon Jeong","orcid":"https://orcid.org/0000-0002-4058-5774"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jongheon Jeong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109878326","display_name":"Yeongjae Cho","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yeongjae Cho","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5017305201","display_name":"Sungzoon Cho","orcid":"https://orcid.org/0000-0002-1695-1973"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sungzoon Cho","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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":0.15769942,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"26642","last_page":"26657"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9997000098228455,"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":0.9997000098228455,"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":0.9635000228881836,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9574000239372253,"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":0.9688446521759033},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7423170208930969},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.679335355758667},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.5648496747016907},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.4894196093082428},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.47448280453681946},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.35708484053611755},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.0861939787864685}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.9688446521759033},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7423170208930969},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.679335355758667},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.5648496747016907},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4894196093082428},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47448280453681946},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.35708484053611755},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0861939787864685},{"id":"https://openalex.org/C21547014","wikidata":"https://www.wikidata.org/wiki/Q1423657","display_name":"Operations management","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.1367","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1367","pdf_url":"https://aclanthology.org/2025.findings-acl.1367.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.1367","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.1367","pdf_url":"https://aclanthology.org/2025.findings-acl.1367.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":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2511445776","display_name":null,"funder_award_id":"2021R1A2C2093785","funder_id":"https://openalex.org/F4320322120","funder_display_name":"National Research Foundation of Korea"},{"id":"https://openalex.org/G30685149","display_name":null,"funder_award_id":"BK21 FOUR","funder_id":"https://openalex.org/F4320320671","funder_display_name":"National Research Foundation"},{"id":"https://openalex.org/G5562307789","display_name":null,"funder_award_id":"BK21 FOUR","funder_id":"https://openalex.org/F4320322120","funder_display_name":"National Research Foundation of Korea"}],"funders":[{"id":"https://openalex.org/F4320320671","display_name":"National Research Foundation","ror":"https://ror.org/05s0g1g46"},{"id":"https://openalex.org/F4320322120","display_name":"National Research Foundation of Korea","ror":"https://ror.org/013aysd81"},{"id":"https://openalex.org/F4320328359","display_name":"Ministry of Science and ICT, South Korea","ror":"https://ror.org/01wpjm123"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412944495.pdf","grobid_xml":"https://content.openalex.org/works/W4412944495.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4210416330","https://openalex.org/W4388002133","https://openalex.org/W3164229987","https://openalex.org/W3215212336","https://openalex.org/W4290852288","https://openalex.org/W3217388757","https://openalex.org/W3122720459","https://openalex.org/W4298897568","https://openalex.org/W1938708284","https://openalex.org/W4380190185"],"abstract_inverted_index":{"Recent":[0],"advances":[1],"in":[2,14],"large":[3],"language":[4],"models":[5,8],"and":[6,28,58,82],"vision-language":[7,77],"have":[9],"led":[10],"to":[11,73],"growing":[12],"interest":[13],"explainable":[15],"evaluation":[16,45,71],"metrics":[17,22],"for":[18,79],"image":[19],"captioning.However,":[20],"these":[21],"generate":[23],"explanations":[24,35,50,95],"without":[25],"standardized":[26],"criteria,":[27],"the":[29,33],"overall":[30],"quality":[31],"of":[32,63],"generated":[34],"remains":[36],"unverified.In":[37],"this":[38],"paper,":[39],"we":[40,67],"propose":[41],"EXPERT,":[42],"a":[43,69,76],"reference-free":[44],"metric":[46],"that":[47],"provides":[48],"structured":[49,65],"based":[51],"on":[52,88],"three":[53],"fundamental":[54],"criteria:":[55],"fluency,":[56],"relevance,":[57],"descriptiveness.By":[59],"constructing":[60],"large-scale":[61],"datasets":[62,90],"highquality":[64],"explanations,":[66],"develop":[68],"two-stage":[70],"template":[72],"effectively":[74],"supervise":[75],"model":[78],"both":[80],"scoring":[81],"explanation":[83],"generation.EXPERT":[84],"achieves":[85],"state-of-the-art":[86],"results":[87],"benchmark":[89],"while":[91],"providing":[92],"significantly":[93],"higherquality":[94],"than":[96],"existing":[97],"metrics,":[98],"as":[99],"validated":[100],"through":[101],"comprehensive":[102],"human":[103],"evaluation.":[104]},"counts_by_year":[],"updated_date":"2026-07-07T14:30:12.667765","created_date":"2025-10-10T00:00:00"}