{"id":"W7077860310","doi":"10.48448/6za3-aa95","title":"ALPACA AGAINST VICUNA: Using LLMs to Uncover Memorization of LLMs","year":2025,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Geochemistry and Geologic Mapping","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Windsor; Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Memorization; Training (meteorology); Training set; Base (topology); Iterative learning control; Factor (programming language)","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":[0.002567414,0.001244585,0.0007420382,0.0005481296,0.0004574441,0.001514701,0.001469076,0.001178353,0.002451583],"category_scores_gemma":[0.02260243,0.0003648107,0.0004667157,0.000266874,0.001289128,0.003082363,0.002572909,0.002384981,0.001227144],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007907709,"about_ca_system_score_gemma":0.001230192,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001633585,"about_ca_topic_score_gemma":0.002160236,"domain_scores_codex":[0.9978001,0.0009208618,0.00009966549,0.0005372636,0.0004252444,0.0002168418],"domain_scores_gemma":[0.9905791,0.00475587,0.0009337213,0.002779998,0.0006264671,0.000324925],"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":[0.002917046,0.0007414629,0.03744886,0.0006600823,0.0002777918,0.001148361,0.002504417,0.2888005,0.06813233,0.02098561,0.01319279,0.5631908],"study_design_scores_gemma":[0.00005606003,0.0006706517,0.00231398,0.00005015202,0.00004493043,0.0002915431,0.0003279628,0.9443246,0.03285278,0.01370174,0.005316903,0.00004877748],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4866433,0.0008246708,0.4795053,0.001537073,0.0002470824,0.0002683345,0.0004428138,0.02345075,0.007080561],"genre_scores_gemma":[0.9394043,0.00009267356,0.05724998,0.0003286559,0.00003454642,0.00008757545,0.0003006007,0.0003657851,0.002135903],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002567414,"threshold_uncertainty_score":0.01357794,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01544067078710087,"score_gpt":0.269653776102785,"score_spread":0.2542131053156841,"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."}}