{"id":"W4385573486","doi":"10.18653/v1/2022.findings-emnlp.191","title":"Partially-Random Initialization: A Smoking Gun for Binarization Hypothesis of BERT","year":2022,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Huawei Technologies (Canada)","funders":"","keywords":"Computer science; Initialization; Embedding; Encoder; Task (project management); Set (abstract data type); Artificial intelligence; Enhanced Data Rates for GSM Evolution; Binary number; Edge device; Machine learning; Programming language; Operating system","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":[0.0003096925,0.00005343801,0.0001022511,0.00007779119,0.000163124,0.00003246633,0.000358312,0.00001688235,0.00009801765],"category_scores_gemma":[0.00008130187,0.00005465483,0.00004667275,0.0002582556,0.000008354896,0.000171692,0.0001844382,0.00002970655,5.696454e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002564652,"about_ca_system_score_gemma":0.00005790091,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002061652,"about_ca_topic_score_gemma":0.000004901723,"domain_scores_codex":[0.9992121,0.00005514741,0.0002229215,0.0001880567,0.0002112416,0.0001104786],"domain_scores_gemma":[0.9994543,0.0001219465,0.00008670109,0.0002520282,0.00006433615,0.00002073814],"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":[0.00006411585,0.0001735257,0.002027486,0.00007179737,0.00005653361,0.000003472578,0.004379815,0.1787124,0.001581408,0.6683155,0.0008664542,0.1437475],"study_design_scores_gemma":[0.0009190053,0.00008090677,0.0001170962,0.000006746891,0.00001040256,0.000003942055,0.00005786277,0.9711159,0.00302519,0.01483662,0.009716556,0.0001097953],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005804888,0.00002534499,0.99187,0.0007412418,0.0002567873,0.0002095692,0.00000217653,0.00007438978,0.001015618],"genre_scores_gemma":[0.9194789,0.000001981203,0.07987978,0.0003618229,0.00004070887,0.00005950606,0.000003741043,0.000006082865,0.0001674573],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9136741,"threshold_uncertainty_score":0.2228759,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04659736887388309,"score_gpt":0.2515191874708972,"score_spread":0.2049218185970141,"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."}}