{"id":"https://openalex.org/W7204146434","doi":"https://doi.org/10.18653/2026.emnlp-main.2117","title":"Re3Cap: Retrieval-Guided Refinement for Image Captioning Enhancement via Reinforcement Learning","display_name":"Re3Cap: Retrieval-Guided Refinement for Image Captioning Enhancement via Reinforcement Learning","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7204146434","doi":"https://doi.org/10.18653/2026.emnlp-main.2117"},"language":"en","primary_location":{"id":"doi:10.18653/2026.emnlp-main.2117","doi":"https://doi.org/10.18653/2026.emnlp-main.2117","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.2117","pdf_url":"https://aclanthology.org/2026.emnlp-main.2117.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"bronze","oa_url":"https://aclanthology.org/2026.emnlp-main.2117.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5147980390","display_name":"Haonan Jia","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Haonan Jia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5143600282","display_name":"Shichao Dong","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shichao Dong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127957820","display_name":"Zenghui Sun","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zenghui Sun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125510291","display_name":"Jiawen Zheng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405261251","display_name":"The Hong Kong University of Science and Technology (Guangzhou)","ror":"https://ror.org/050h0vm43","country_code":"CN","type":"education","lineage":["https://openalex.org/I4405261251"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiawen Zheng","raw_affiliation_strings":["The Hong Kong University of Science and Technology (Guangzhou )"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Hong Kong University of Science and Technology (Guangzhou )","institution_ids":["https://openalex.org/I4405261251"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5117373548","display_name":"Ziqi Miao","orcid":"https://orcid.org/0009-0002-9087-7879","observed_orcids":["https://orcid.org/0009-0002-9087-7879"]},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ziqi Miao","raw_affiliation_strings":["Shanghai Artificial Intelligence Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Artificial Intelligence Laboratory","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062214569","display_name":"Gege Shi","orcid":"https://orcid.org/0009-0003-1526-9002","observed_orcids":["https://orcid.org/0009-0003-1526-9002"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gege Shi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113554811","display_name":"Qiuyu Zhao","orcid":"https://orcid.org/0009-0006-5458-4787","observed_orcids":["https://orcid.org/0009-0006-5458-4787"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qiuyu Zhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053940023","display_name":"Jinsong Lan","orcid":"https://orcid.org/0009-0000-6890-4960","observed_orcids":["https://orcid.org/0009-0000-6890-4960"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinsong Lan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128219714","display_name":"Xiaoyong Zhu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaoyong Zhu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5127981713","display_name":"Bo Zheng","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bo Zheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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":0.68938986,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"48154","last_page":"48169"},"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/T10824","display_name":"Image Retrieval and Classification Techniques","score":0.0,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.0,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/image-captioning","display_name":"image captioning","score":0.9819999933242798},{"id":"https://openalex.org/keywords/retrieval-augmented-generation","display_name":"retrieval-augmented generation","score":0.7580000162124634},{"id":"https://openalex.org/keywords/multimodal-retrieval","display_name":"multimodal retrieval","score":0.8159999847412109},{"id":"https://openalex.org/keywords/large-vision-language-models","display_name":"large vision-language models","score":0.9430000185966492},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"reinforcement learning","score":0.6370000243186951},{"id":"https://openalex.org/keywords/hallucination-detection","display_name":"hallucination detection","score":0.9340000152587891}],"concepts":[{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.9850000143051147},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7754999995231628},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7321000099182129},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6912000179290771},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.6741999983787537},{"id":"https://openalex.org/C25343380","wikidata":"https://www.wikidata.org/wiki/Q277521","display_name":"Relation (database)","level":2,"score":0.5264999866485596},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.427700012922287},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.4032000005245209},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3946000039577484},{"id":"https://openalex.org/C1667742","wikidata":"https://www.wikidata.org/wiki/Q10927554","display_name":"Image retrieval","level":3,"score":0.38519999384880066},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.35519999265670776},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.3517000079154968},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3476000130176544},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.31369999051094055},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.31299999356269836},{"id":"https://openalex.org/C55020928","wikidata":"https://www.wikidata.org/wiki/Q3813865","display_name":"Image quality","level":3,"score":0.2928999960422516},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2508000135421753}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/2026.emnlp-main.2117","doi":"https://doi.org/10.18653/2026.emnlp-main.2117","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.2117","pdf_url":"https://aclanthology.org/2026.emnlp-main.2117.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2608.21305","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.21305","pdf_url":"https://arxiv.org/pdf/2608.21305","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.21305","doi":"https://doi.org/10.48550/arxiv.2608.21305","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.21305","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.18653/2026.emnlp-main.2117","doi":"https://doi.org/10.18653/2026.emnlp-main.2117","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.2117","pdf_url":"https://aclanthology.org/2026.emnlp-main.2117.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204146434.pdf","grobid_xml":"https://content.openalex.org/works/W7204146434.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7083704284","https://openalex.org/W7160847285","https://openalex.org/W7164080106","https://openalex.org/W4393937467","https://openalex.org/W4415353920","https://openalex.org/W7134898772","https://openalex.org/W4402916921","https://openalex.org/W7131634831","https://openalex.org/W2903928149","https://openalex.org/W4415192993"],"abstract_inverted_index":{"Reinforcement":[0],"Learning":[1],"(RL)":[2],"has":[3],"demonstrated":[4],"significant":[5],"gains":[6],"in":[7,15,99,116,135],"image":[8,75,100,117],"captioning,":[9,118],"yet":[10],"it":[11],"is":[12],"still":[13],"limited":[14],"encouraging":[16],"Large":[17],"Vision-Language":[18],"Models":[19],"(LVLMs)":[20],"to":[21,28,103],"explore":[22],"novel":[23],"reasoning":[24,50,71,137],"strategies.This":[25],"limitation":[26],"leads":[27],"a":[29,69],"performance":[30],"gap":[31],"between":[32],"RL":[33],"and":[34,88,97,106],"Supervised":[35,122],"Fine-Tuning":[36],"(SFT).In":[37],"this":[38,56,93],"paper,":[39],"we":[40,58],"argue":[41],"that":[42,73],"multi-modal":[43],"retrieval":[44,79],"can":[45],"serve":[46],"as":[47],"an":[48,130],"effective":[49],"signal":[51],"for":[52,63],"caption":[53],"refinement.Based":[54],"on":[55,80,138],"insight,":[57],"present":[59],"the":[60,111,139],"Retrieval-Guided":[61],"Refinement":[62,85],"Image":[64],"Captioning":[65],"(Re":[66],"3":[67,125],"Cap),":[68],"retrieval-guided":[70],"strategy":[72,94],"enhances":[74],"captioning":[76],"by":[77,83],"performing":[78],"support":[81],"sets.Instantiated":[82],"Caption":[84,89],"Suggester":[86],"(CRS)":[87],"Quality":[90],"Assessor":[91],"(CQA),":[92],"identifies":[95],"hallucinations":[96],"omissions":[98],"captions,":[101],"leading":[102],"more":[104],"accurate":[105],"detailed":[107],"descriptions.Extensive":[108],"experiments":[109],"demonstrate":[110],"superiority":[112],"of":[113,133],"our":[114],"method":[115],"even":[119],"compared":[120],"with":[121,129],"Fine-Tuning.Especially,":[123],"Re":[124],"Cap":[126],"outperforms":[127],"GRPO":[128],"average":[131],"improvement":[132],"8.64%":[134],"relation":[136],"COCO-LN500":[140],"benchmark.\ud835\udc60":[141],"!\"":[142]},"counts_by_year":[],"updated_date":"2026-10-31T12:09:11.096373","created_date":"2026-08-25T00:00:00"}