{"id":"W4387324024","doi":"10.48550/arxiv.2310.00893","title":"Engineering the Neural Collapse Geometry of Supervised-Contrastive Loss","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Visual Attention and Saliency Detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; King Abdullah University of Science and Technology; Universities Space Research Association; National Science Foundation","keywords":"Embedding; Limiting; Artificial intelligence; Computer science; Classifier (UML); Entropy (arrow of time); Benchmark (surveying); Feature (linguistics); Geometry; Cross entropy; Artificial neural network; Feature vector; Pattern recognition (psychology); Machine learning; Mathematics; Algorithm; Engineering; Mechanical engineering","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004099088,0.001243729,0.001026661,0.0008474992,0.0004686271,0.001523844,0.001677035,0.001650881,0.002196788],"category_scores_gemma":[0.01658469,0.0005165007,0.0005659919,0.000444783,0.002449052,0.003587472,0.003746399,0.003002371,0.000668797],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001851331,"about_ca_system_score_gemma":0.001143904,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00111563,"about_ca_topic_score_gemma":0.001288391,"domain_scores_codex":[0.9984964,0.0006089437,0.00006401342,0.0003331754,0.0003790162,0.0001184512],"domain_scores_gemma":[0.9960645,0.002065732,0.0004108332,0.0006353257,0.0005559639,0.000267631],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005316508,0.0003049145,0.004255578,0.0001964937,0.0000938321,0.0003616168,0.0002160174,0.6795164,0.0176594,0.1503228,0.007789098,0.1387523],"study_design_scores_gemma":[0.00001606033,0.0001283196,0.0003156938,0.00001302245,0.000005684786,0.0000772208,0.00001306077,0.9611642,0.002536837,0.03510914,0.0006095627,0.00001111644],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06400502,0.0002773743,0.9300452,0.0008346699,0.00005518338,0.00008635143,0.0001268947,0.0006414912,0.003927711],"genre_scores_gemma":[0.8531579,0.0002709879,0.1403145,0.0005260332,0.0001181065,0.0002704922,0.0003416799,0.0003070942,0.004693306],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004099088,"threshold_uncertainty_score":0.02167827,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06787373052823772,"score_gpt":0.1952296738075231,"score_spread":0.1273559432792854,"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."}}