{"id":"W2962849532","doi":"","title":"Teaching Machines to Describe Images with Natural Language Feedback","year":2017,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":38,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Closed captioning; Computer science; Natural language; Focus (optics); Artificial intelligence; Sentence; Point (geometry); Natural (archaeology); Quality (philosophy); Phrase; Natural language processing; SIGNAL (programming language); Robot; Human–computer interaction; Machine learning; Image (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":"codex-gemma-dda1882f352a","candidate_categories":["sts","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003322446,0.0001916997,0.0001761726,0.0001534093,0.001382242,0.003866208,0.001374958,0.00004288179,0.000001328125],"category_scores_gemma":[0.0001799532,0.0001425501,0.00002993991,0.0001312528,0.00004108064,0.00645049,0.0002451818,0.0003543471,0.0001243628],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005004885,"about_ca_system_score_gemma":0.00004610546,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00243764,"about_ca_topic_score_gemma":0.00001754775,"domain_scores_codex":[0.9987383,0.00005429094,0.0003352173,0.0002234065,0.0003732597,0.0002754882],"domain_scores_gemma":[0.9984029,0.00003363966,0.0004766326,0.0007854183,0.0001860165,0.0001154278],"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":[0.00002509135,0.00002930847,0.006846467,0.0004537156,0.00001624234,0.0000059815,0.01595459,0.01220714,0.002265054,0.004327306,0.0009602955,0.9569088],"study_design_scores_gemma":[0.0003451883,0.00004760069,0.03526272,0.0001761072,0.00000558333,0.00010795,0.0003383776,0.9607921,0.0002278743,0.00001559467,0.002383623,0.000297265],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4898584,0.0002528053,0.4823173,0.01024906,0.0008288963,0.001090136,0.00001283825,0.001416995,0.01397356],"genre_scores_gemma":[0.9788544,2.413934e-7,0.01997457,0.0004714189,0.0001174468,0.0000799722,0.00001237648,0.00001133977,0.0004782702],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9566116,"threshold_uncertainty_score":0.9999178,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01093533391580301,"score_gpt":0.2877543883028231,"score_spread":0.2768190543870201,"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."}}