{"id":"W4399554391","doi":"10.48550/arxiv.2406.06443","title":"LLM Dataset Inference: Did you train on my dataset?","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Defense Advanced Research Projects Agency; Government of Canada; Canadian Institute for Advanced Research; Alfred P. Sloan Foundation","keywords":"Inference; Computer science; Artificial intelligence; Machine learning; Data mining","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0141876,0.001959072,0.001183074,0.002397579,0.001526793,0.003193454,0.004507038,0.003707093,0.00713436],"category_scores_gemma":[0.05872321,0.0009414698,0.002031403,0.001905361,0.001257487,0.00678779,0.002719253,0.005974955,0.01020627],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002290354,"about_ca_system_score_gemma":0.002619286,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01272219,"about_ca_topic_score_gemma":0.02400694,"domain_scores_codex":[0.994001,0.001974656,0.0004118805,0.001880336,0.001238106,0.0004939517],"domain_scores_gemma":[0.9843547,0.006144714,0.0006464577,0.005554358,0.002717756,0.000581987],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000769657,0.0007704627,0.04739794,0.0006926421,0.0006663139,0.0003460124,0.0003498436,0.03184507,0.003345321,0.00783944,0.672342,0.2336352],"study_design_scores_gemma":[0.001024717,0.0008344475,0.02709767,0.0008886846,0.0003182039,0.001059745,0.001018785,0.5946565,0.01858846,0.04970387,0.3045078,0.0003010744],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.185436,0.007072792,0.3383215,0.04948738,0.008304728,0.002114716,0.2429148,0.1360141,0.030334],"genre_scores_gemma":[0.2582075,0.0009915062,0.3460138,0.01159079,0.001002777,0.001536382,0.3663349,0.006064288,0.008258161],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0141876,"threshold_uncertainty_score":0.07503211,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1126358343808321,"score_gpt":0.2624317010573041,"score_spread":0.149795866676472,"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."}}