{"id":"W4411176006","doi":"10.1016/j.nlp.2025.100159","title":"Next-generation image captioning: A survey of methodologies and emerging challenges from transformers to Multimodal Large Language Models","year":2025,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Closed captioning; Computer science; Transformer; Natural language processing; Artificial intelligence; Image (mathematics); Engineering; Electrical engineering","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001333279,0.0001802716,0.0002734283,0.0002949322,0.0003396052,0.000349073,0.0004852718,0.00008957508,0.000005958382],"category_scores_gemma":[0.0006287405,0.000153644,0.00005412213,0.0004377197,0.00003877351,0.0009760634,0.0001324616,0.0005530218,0.000001194344],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005261588,"about_ca_system_score_gemma":0.0001136237,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0015696,"about_ca_topic_score_gemma":0.0004846693,"domain_scores_codex":[0.9983776,0.0003441494,0.0003580232,0.0003682454,0.0002654352,0.0002865065],"domain_scores_gemma":[0.9989703,0.0002816457,0.0001934501,0.0001985206,0.000267955,0.00008817682],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002665587,0.00003644223,0.0003352919,0.0000558355,0.00003844604,0.000007667559,0.0287252,0.00128723,0.1492496,0.0004874309,0.00002012234,0.81973],"study_design_scores_gemma":[0.0005892712,0.0000300004,0.03068471,0.0001500087,0.00002322757,0.00002108556,0.003043111,0.957644,0.007010986,0.0005701382,0.00001119952,0.000222228],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.362071,0.02323296,0.6128168,0.001457518,0.0001136883,0.0001100501,0.00001165097,0.00007837762,0.0001080101],"genre_scores_gemma":[0.6997854,0.00009196009,0.2999023,0.0001168452,0.00005051312,0.000007910132,0.0000138664,0.000008405426,0.00002289935],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9563568,"threshold_uncertainty_score":0.6265422,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0630583087593844,"score_gpt":0.3692293337405086,"score_spread":0.3061710249811243,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}