{"id":"W4411950007","doi":"10.1109/tit.2025.3584013","title":"Next-Token Prediction Capacity: General Upper Bounds and a Lower Bound for Transformers","year":2025,"lang":"en","type":"article","venue":"IEEE Transactions on Information Theory","topic":"Computational Physics and Python Applications","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Upper and lower bounds; Computer science; Security token; Capacity planning; Mathematics; Algorithm; Computer network","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0002826171,0.0001371559,0.0001129887,0.0002927347,0.0005313227,0.0003622828,0.0001908074,0.00007188198,0.00001553777],"category_scores_gemma":[0.000003655572,0.0001392384,0.0001072949,0.0003809114,0.0000777806,0.002091702,0.00000212268,0.0001352933,0.00002263649],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007216226,"about_ca_system_score_gemma":0.000110731,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001060561,"about_ca_topic_score_gemma":0.000003366821,"domain_scores_codex":[0.9991491,0.00003233786,0.0003149974,0.000170071,0.000163022,0.0001703999],"domain_scores_gemma":[0.999306,0.0001996758,0.00006464527,0.0002175475,0.0001479578,0.00006423362],"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":[0.00007100939,0.00007812124,0.000002016534,0.00004349505,0.00006686702,3.831925e-8,0.001207715,0.008506979,0.000250207,0.7190679,0.0006532041,0.2700524],"study_design_scores_gemma":[0.001553512,0.0001940875,0.0004867044,0.00004523393,0.00006702889,0.000009380824,0.0001988868,0.5260065,0.00430698,0.3912182,0.07557482,0.0003386439],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009367785,0.00001278367,0.9841825,0.0010004,0.0007054251,0.0005037216,0.0001164473,0.0001580151,0.003952903],"genre_scores_gemma":[0.9856969,0.00003321041,0.01118281,0.001972373,0.00004067002,0.0004179772,0.00002920369,0.000007374665,0.0006195247],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9763291,"threshold_uncertainty_score":0.5677977,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01034835890367836,"score_gpt":0.2351723872306748,"score_spread":0.2248240283269964,"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."}}