{"id":"W4410007609","doi":"10.1038/s42256-025-01033-7","title":"Lossless data compression by large models","year":2025,"lang":"en","type":"article","venue":"Nature Machine Intelligence","topic":"Algorithms and Data Compression","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Lossless compression; Compression (physics); Computer science; Data compression; Materials science; Algorithm; Composite material","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":[0.001001099,0.0006673971,0.0008393795,0.0008013926,0.0003675747,0.001209739,0.001057125,0.001052334,0.002509517],"category_scores_gemma":[0.005935116,0.000427169,0.000621497,0.001219282,0.000966026,0.003530935,0.00135572,0.002011273,0.001293053],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006765382,"about_ca_system_score_gemma":0.0005354142,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001300804,"about_ca_topic_score_gemma":0.001486674,"domain_scores_codex":[0.9992657,0.0001983944,0.00003995082,0.0001112572,0.0003312654,0.00005344435],"domain_scores_gemma":[0.9978096,0.001109568,0.0001035342,0.0007539841,0.0001802155,0.00004314234],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006623317,0.0001867007,0.001177584,0.0002896359,0.0001127677,0.0003007894,0.0001192817,0.4823742,0.02190203,0.06091945,0.01327254,0.4186828],"study_design_scores_gemma":[0.00000927728,0.00002560982,0.0001170107,0.00001118509,0.00000835789,0.00006454255,0.000009104238,0.9768463,0.00472234,0.01684389,0.001335872,0.000006464867],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0217582,0.001235254,0.9716042,0.001354765,0.0002404287,0.00004841395,0.0003312079,0.001419671,0.002007835],"genre_scores_gemma":[0.7011406,0.002313524,0.280834,0.0008345205,0.0004915086,0.0002396492,0.00132316,0.0003727642,0.01245021],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002509517,"threshold_uncertainty_score":0.008395135,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01614885772911826,"score_gpt":0.3229046710219013,"score_spread":0.3067558132927831,"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."}}