{"id":"W4389520158","doi":"10.18653/v1/2023.findings-emnlp.404","title":"NASH: A Simple Unified Framework of Structured Pruning for Accelerating Encoder-Decoder Language Models","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"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; Encoder; Speedup; Inference; Pruning; Language model; Algorithm; Artificial intelligence; Parallel computing","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":[0.000322565,0.0001333383,0.0002118263,0.0001326967,0.0001219225,0.0001079286,0.0007282601,0.0001080349,0.0000234155],"category_scores_gemma":[0.0001325374,0.0001203554,0.00007667019,0.0004783804,0.00001473438,0.0004112937,0.0003048652,0.0001363738,0.000003881495],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001927901,"about_ca_system_score_gemma":0.00006859338,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006638307,"about_ca_topic_score_gemma":0.0000364848,"domain_scores_codex":[0.9986643,0.00003132964,0.000331195,0.0003819338,0.0002386121,0.0003526121],"domain_scores_gemma":[0.9987819,0.0003483633,0.0001051641,0.0006091379,0.00009244983,0.00006297089],"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":[0.000006962538,0.00001499381,0.0001274423,0.00009703379,0.00003552372,0.000005752443,0.0158456,0.2761031,0.009763595,0.657684,0.0004518933,0.03986414],"study_design_scores_gemma":[0.0001694524,0.00001778354,0.00003586187,0.00002083996,0.000002913018,0.000001112522,0.0005999485,0.8259571,0.005586615,0.167458,0.00003282134,0.0001176066],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1492977,0.00002649187,0.8487168,0.0002191076,0.0001533541,0.0002459766,0.000004684404,0.0003735273,0.0009624041],"genre_scores_gemma":[0.5901874,0.000001320387,0.4093873,0.000132017,0.00005017478,0.00001732672,0.000004719007,0.00001057344,0.0002091633],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.549854,"threshold_uncertainty_score":0.4907952,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06707218725630042,"score_gpt":0.3174647955264347,"score_spread":0.2503926082701343,"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."}}