{"id":"W4402821735","doi":"10.1101/2024.09.23.614603","title":"Protein Language Models: Is Scaling Necessary?","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Topic Modeling","field":"Computer Science","cited_by":30,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; Polytechnique Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Scaling; Computer science; Mathematics; Geometry","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":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0008254375,0.0007044391,0.0006062041,0.000426348,0.0001631702,0.00123053,0.002621069,0.0006236758,0.00002366222],"category_scores_gemma":[0.00007207075,0.0007481118,0.0002613726,0.0007485442,0.00006375364,0.0004714127,0.004124075,0.00156906,0.000267612],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002990379,"about_ca_system_score_gemma":0.0008147606,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000137606,"about_ca_topic_score_gemma":0.000001396759,"domain_scores_codex":[0.9955483,0.0001279366,0.0006926506,0.002063758,0.0007637839,0.0008035469],"domain_scores_gemma":[0.99588,0.00003787639,0.0002926568,0.003167332,0.0002962168,0.0003259208],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002457794,0.0003911325,0.0001479297,0.005838884,0.0007722654,0.001575132,0.00140886,0.01186101,0.763238,0.2114037,0.00305843,0.0002800761],"study_design_scores_gemma":[0.000274098,0.00002206395,0.00008298387,0.001726811,0.00007009153,4.020862e-8,0.000008599261,0.7477239,0.2465935,0.0008156637,0.001291964,0.001390278],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4504142,0.007626909,0.5332685,0.001921124,0.002465116,0.001221159,0.0001580523,0.002810455,0.0001145123],"genre_scores_gemma":[0.8889127,0.00004947726,0.1094741,0.0004869668,0.0006420886,0.000260886,1.216973e-7,0.0001348599,0.00003887983],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7358629,"threshold_uncertainty_score":0.9998063,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02002412734202891,"score_gpt":0.231310756102787,"score_spread":0.2112866287607581,"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."}}