{"id":"W3162126217","doi":"10.48550/arxiv.2105.08665","title":"A multimodal deep learning framework for scalable content based visual media retrieval","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Image and Video Retrieval Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Scalability; Modular design; Deep learning; Metric (unit); Artificial intelligence; Content-based image retrieval; Image retrieval; Information retrieval; Image (mathematics); Database","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"],"consensus_categories":[],"category_scores_codex":[0.0004436185,0.000453998,0.0005996377,0.0002765238,0.0002931825,0.0003094784,0.001531953,0.0006213156,0.0000402978],"category_scores_gemma":[0.001371434,0.0005319485,0.0004760268,0.0008604662,0.000141482,0.0006310854,0.001663935,0.001346551,0.00001428759],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003203326,"about_ca_system_score_gemma":0.0003302799,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004240218,"about_ca_topic_score_gemma":0.0000101243,"domain_scores_codex":[0.9970798,0.0002121108,0.0003237487,0.001570761,0.00019976,0.0006138455],"domain_scores_gemma":[0.996683,0.001019421,0.0003786162,0.0009726529,0.0006751596,0.0002711621],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003500381,0.002823319,0.01622421,0.002263817,0.001279799,0.005614269,0.003303941,0.4132841,0.009618447,0.4434026,0.0003002464,0.09838482],"study_design_scores_gemma":[0.0009281184,0.0002470092,0.0004077076,0.0004102702,0.00008681282,0.000003632436,0.0002859032,0.9398385,0.02991979,0.02665046,0.0004199607,0.0008018105],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05352406,0.0002557617,0.9441987,0.00008095565,0.0005626154,0.0005796404,0.000009936083,0.0007023973,0.0000859346],"genre_scores_gemma":[0.7326825,0.0001738681,0.2664469,0.00018277,0.0001122344,0.000003333086,0.00005250647,0.0000350784,0.0003108102],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6791585,"threshold_uncertainty_score":0.9997132,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08949241081933848,"score_gpt":0.2454015779684627,"score_spread":0.1559091671491243,"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."}}