{"id":"W4404782398","doi":"10.18653/v1/2024.emnlp-main.606","title":"Stable Language Model Pre-training by Reducing Embedding Variability","year":2024,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"Institute for Information and Communications Technology Promotion; Ministry of Science and ICT, South Korea; Korea Advanced Institute of Science and Technology","keywords":"Computer science; Training (meteorology); Embedding; Language model; Artificial intelligence; Natural language processing","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":[],"consensus_categories":[],"category_scores_codex":[0.0009291893,0.0001143188,0.0001222575,0.00005476229,0.00008231216,0.0003974177,0.0005017762,0.00005346483,0.00005669765],"category_scores_gemma":[0.00007047502,0.000102479,0.00004604556,0.0002008144,0.00001265719,0.0007535882,0.0002579569,0.0001782264,0.00001784354],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009394747,"about_ca_system_score_gemma":0.0001329981,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002250305,"about_ca_topic_score_gemma":0.000003095749,"domain_scores_codex":[0.9986193,0.00004290155,0.0002125493,0.0005794493,0.0002172049,0.0003285515],"domain_scores_gemma":[0.9991722,0.000130255,0.00001889185,0.0005770687,0.00002083328,0.00008074136],"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":[0.000001857728,0.00003756323,0.00004295732,0.0001562908,0.00003068124,0.00002308307,0.04349254,0.4775438,0.04103439,0.2886856,0.004018221,0.144933],"study_design_scores_gemma":[0.00003648053,0.000005542166,0.000001826656,0.00004406308,0.000003119872,0.000005801891,0.000186175,0.9913862,0.002343727,0.00554802,0.0003106757,0.0001283841],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05860855,0.0002039836,0.9213287,0.0003138887,0.0002921326,0.00008705071,0.000002884138,0.0008223146,0.01834047],"genre_scores_gemma":[0.7194446,0.000001812331,0.276767,0.0001004024,0.00005142445,0.00000773155,0.000001401083,0.000008692791,0.003616902],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.660836,"threshold_uncertainty_score":0.417897,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02390127958440878,"score_gpt":0.2959054770229754,"score_spread":0.2720041974385666,"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."}}