{"id":"W4385574133","doi":"10.18653/v1/2022.findings-emnlp.240","title":"AlphaTuning: Quantization-Aware Parameter-Efficient Adaptation of Large-Scale Pre-Trained Language Models","year":2022,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Computer science; Quantization (signal processing); Adaptation (eye); Language model; Scale (ratio); Artificial intelligence; Natural language processing; Speech recognition; Psychology; Algorithm; Physics; Quantum mechanics; Neuroscience","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00147786,0.001165448,0.001015552,0.0007727705,0.0004055452,0.0009658837,0.001947994,0.001064825,0.003509433],"category_scores_gemma":[0.007215063,0.0006451022,0.0006353814,0.0008925586,0.0004917282,0.001951615,0.001831705,0.002145612,0.002403725],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006062436,"about_ca_system_score_gemma":0.0008919886,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007455328,"about_ca_topic_score_gemma":0.01115105,"domain_scores_codex":[0.9993319,0.0002125231,0.00004976417,0.0002256772,0.0001139464,0.00006614559],"domain_scores_gemma":[0.9977105,0.001169491,0.000088156,0.0005166655,0.0004134692,0.0001017827],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000638967,0.0002941131,0.001633272,0.0001740438,0.0001654997,0.0002003362,0.0003002018,0.1357495,0.03700552,0.001960125,0.0178178,0.8040606],"study_design_scores_gemma":[0.00004593395,0.00006570848,0.0005950701,0.00001512322,0.00003465349,0.00005584041,0.00004717632,0.9844289,0.01009077,0.002404022,0.002193192,0.00002355681],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06477676,0.003003792,0.9050478,0.0003353472,0.0007788535,0.0001545981,0.000669206,0.02120191,0.004031772],"genre_scores_gemma":[0.6582957,0.001055703,0.3282899,0.0003935962,0.0002307449,0.0003449097,0.002823938,0.002423339,0.006142207],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007455328,"threshold_uncertainty_score":0.01482385,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02893176651273319,"score_gpt":0.2697405083380179,"score_spread":0.2408087418252847,"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."}}