{"id":"W2963299217","doi":"","title":"A Neural Compositional Paradigm for Image Captioning","year":2018,"lang":"en","type":"article","venue":"The HKU Scholars Hub (University of Hong Kong)","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Closed captioning; Computer science; Semantics (computer science); Generalization; Natural language processing; Artificial intelligence; Syntax; Representation (politics); Image (mathematics); Programming language; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009572725,0.0008981592,0.0004754021,0.0007284245,0.0006329531,0.0009322952,0.001762701,0.001357747,0.004508812],"category_scores_gemma":[0.003008685,0.0004061788,0.001130743,0.0007595345,0.00122641,0.002447386,0.001422018,0.001740929,0.001175711],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006930669,"about_ca_system_score_gemma":0.0007149016,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001669025,"about_ca_topic_score_gemma":0.002550569,"domain_scores_codex":[0.9995209,0.0001481523,0.00002624234,0.0001796048,0.00009003973,0.00003504915],"domain_scores_gemma":[0.9994604,0.0002221094,0.0000435949,0.0001246427,0.0001180317,0.00003119735],"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":[0.0001957287,0.0001734813,0.0004142384,0.0005127179,0.0001111088,0.0003317971,0.0005783704,0.1766132,0.06113108,0.1925826,0.01056243,0.5567932],"study_design_scores_gemma":[0.00001716032,0.0000786664,0.0001490277,0.00002603859,0.00002886751,0.0001600326,0.00003788458,0.8932787,0.01128026,0.08661977,0.008299908,0.00002368951],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003868385,0.0001874218,0.9916746,0.0001614252,0.00006242489,0.00006812032,0.00006879564,0.0006008729,0.00330797],"genre_scores_gemma":[0.2540931,0.0006487569,0.7359983,0.0003939862,0.000223398,0.0004287953,0.0006164403,0.0002606357,0.00733651],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004508812,"threshold_uncertainty_score":0.01508349,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01437617044879723,"score_gpt":0.2468396403155849,"score_spread":0.2324634698667877,"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."}}