{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001081777,0.0009391811,0.0003699527,0.0002611283,0.0002114537,0.0007390174,0.001201071,0.001222296,0.003486216],"category_scores_gemma":[0.009320332,0.0003636059,0.0003864178,0.000259935,0.0009971771,0.002122013,0.0008336997,0.001769566,0.001367244],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004810489,"about_ca_system_score_gemma":0.0004600273,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001265521,"about_ca_topic_score_gemma":0.002557572,"domain_scores_codex":[0.9995632,0.0002154564,0.00001688359,0.0001063777,0.0000734343,0.00002465956],"domain_scores_gemma":[0.9968693,0.002179979,0.000252858,0.0003372068,0.0002880157,0.00007265253],"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":[0.0002160874,0.0003462971,0.00266506,0.0006988556,0.00009992635,0.0002865715,0.001235724,0.3835717,0.06168546,0.05068436,0.01880184,0.4797081],"study_design_scores_gemma":[0.00002313697,0.0001052694,0.0003518741,0.00004019316,0.00001824109,0.0001052604,0.00008217298,0.941195,0.01704113,0.0330693,0.007941335,0.00002713519],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03316235,0.0004076996,0.9558847,0.001385749,0.0001567392,0.0001182426,0.0001348432,0.003208072,0.005541677],"genre_scores_gemma":[0.5895531,0.0008168659,0.3983003,0.0008281455,0.0001715387,0.0002684662,0.0003971767,0.0003492072,0.00931522],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003486216,"threshold_uncertainty_score":0.01166254,"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."}}