{"id":"W2963499204","doi":"","title":"VSE++: Improving Visual-Semantic Embeddings with Hard Negatives.","year":2018,"lang":"en","type":"article","venue":"British Machine Vision Conference","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":230,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Modal; Ranking (information retrieval); Information retrieval; Learning to rank; Artificial intelligence; Simple (philosophy); Image retrieval; Image (mathematics); Machine learning; Data mining; Pattern recognition (psychology)","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":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003694295,0.0002931294,0.0003222998,0.0001604899,0.0005778088,0.00123737,0.00135726,0.00009239544,0.0002801265],"category_scores_gemma":[0.0002279465,0.0003002574,0.00006668489,0.0007357721,0.0003145655,0.0008504343,0.0006379909,0.0004269632,0.0003979857],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004610964,"about_ca_system_score_gemma":0.0001672677,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005798887,"about_ca_topic_score_gemma":0.0006568564,"domain_scores_codex":[0.9973624,0.0001269367,0.0003849529,0.001065767,0.0005393709,0.0005205125],"domain_scores_gemma":[0.998044,0.0001485743,0.0002608821,0.0007912835,0.0005075181,0.0002477865],"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":[0.00002513836,0.0002386063,0.008211326,0.00005392958,0.00002939594,0.00005651906,0.0006470091,0.00004335191,0.01158254,0.01556618,0.0006872564,0.9628587],"study_design_scores_gemma":[0.0008213702,0.0006566258,0.2024756,0.0002719724,0.00001371052,0.0002458565,0.00001948426,0.7900173,0.001544168,0.001395001,0.001980978,0.0005579694],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2423177,0.0000848574,0.7523491,0.001368191,0.0001424522,0.000364636,0.00000825354,0.0006299616,0.002734854],"genre_scores_gemma":[0.9070306,0.00002058191,0.09171885,0.0004589644,0.0001298263,0.00005221535,0.00001191207,0.0000339715,0.0005430511],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9623008,"threshold_uncertainty_score":0.9999449,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01003198278439735,"score_gpt":0.2815968184575066,"score_spread":0.2715648356731092,"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."}}