{"id":"W4402670576","doi":"10.18653/v1/2024.findings-acl.621","title":"Mixture-of-Supernets: Improving Weight-Sharing Supernet Training with Architecture-Routed Mixture-of-Experts","year":2024,"lang":"en","type":"article","venue":"","topic":"Online Learning and Analytics","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Alliance de recherche numérique du Canada; Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Computer science; Training (meteorology); Architecture; Computer architecture; Multimedia","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002917436,0.000316892,0.0004391959,0.0003452236,0.00007499988,0.0002259889,0.001029625,0.0001290986,0.00003559683],"category_scores_gemma":[0.0000651395,0.0002201122,0.0001696382,0.0008610323,0.0001023537,0.0003581543,0.0002639797,0.0004398101,0.000007632114],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003063299,"about_ca_system_score_gemma":0.0002120188,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000220553,"about_ca_topic_score_gemma":0.00005027059,"domain_scores_codex":[0.99776,0.00005461439,0.0004742153,0.0007386025,0.0004937768,0.0004788192],"domain_scores_gemma":[0.9987971,0.0002138658,0.00009652276,0.0006654126,0.0000804606,0.0001465957],"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":[0.00009559373,0.0004079907,0.007283868,0.001589152,0.0009862182,0.0006047487,0.07753139,0.00808236,0.1003552,0.2310175,0.001134939,0.570911],"study_design_scores_gemma":[0.0007030629,0.000654849,0.0002872633,0.001024237,0.0001017682,0.0002642765,0.0009138742,0.9535862,0.02547824,0.004676218,0.0114179,0.0008921017],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2009436,0.002896036,0.7847898,0.003104501,0.0006369242,0.000243983,0.000008902387,0.001008056,0.006368154],"genre_scores_gemma":[0.889204,0.00001537927,0.1089336,0.0001357333,0.0002065705,0.000005004834,0.000009480385,0.00003799906,0.001452269],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9455038,"threshold_uncertainty_score":0.8975916,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01260724651315749,"score_gpt":0.2406474297289617,"score_spread":0.2280401832158042,"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."}}