{"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":"codex-gemma-dda1882f352a","candidate_categories":["open_science"],"consensus_categories":[],"category_scores_codex":[0.0004348457,0.0002450013,0.0002397526,0.0001303336,0.000269485,0.0002447895,0.005954377,0.0002826069,0.00003760592],"category_scores_gemma":[0.000082078,0.00019066,0.00004571112,0.0006482037,0.00004495026,0.001316762,0.004775415,0.00101,0.00003620384],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003175387,"about_ca_system_score_gemma":0.00007234986,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009403581,"about_ca_topic_score_gemma":0.0000176511,"domain_scores_codex":[0.9978213,0.00008662524,0.0003251686,0.0009436883,0.0004396214,0.0003835358],"domain_scores_gemma":[0.9967923,0.0001717466,0.00009855525,0.00271835,0.0001150678,0.0001039682],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002545215,0.0003579322,0.0001400496,0.00005636243,0.00003488067,0.00003727461,0.0001076834,0.0008731496,0.000577431,0.3338642,0.2896139,0.3743116],"study_design_scores_gemma":[0.0001012969,0.00001309037,0.0000446931,0.0001126396,0.000005857604,0.00000507364,0.000007683107,0.8540062,0.004922285,0.02706764,0.1135214,0.0001921466],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00009632098,0.01215541,0.979853,0.001354785,0.001328321,0.0001755963,0.0004016237,0.0002309179,0.004404049],"genre_scores_gemma":[0.9177517,0.0006180032,0.07554677,0.003439308,0.0001111499,0.00001242547,0.0008536353,0.00001882105,0.00164815],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9176554,"threshold_uncertainty_score":0.9994239,"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."}}