{"id":"W4416034840","doi":"10.18653/v1/2025.findings-emnlp.609","title":"How Sampling Affects the Detectability of Machine-written texts: A Comprehensive Study","year":2025,"lang":"en","type":"article","venue":"","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Sampling (signal processing); Noise (video); Measure (data warehouse)","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":[],"consensus_categories":[],"category_scores_codex":[0.0005255954,0.000118206,0.0001958785,0.00007161257,0.0002058391,0.0001407869,0.0008080506,0.0000418021,0.000007675472],"category_scores_gemma":[0.000210333,0.00007401636,0.00007699193,0.000666977,0.00006201164,0.0001570989,0.0004650529,0.0001954299,0.000003436568],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002962952,"about_ca_system_score_gemma":0.00006096352,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006125139,"about_ca_topic_score_gemma":0.0000641937,"domain_scores_codex":[0.9987799,0.0003460101,0.0001697285,0.0003104263,0.0002030683,0.0001908667],"domain_scores_gemma":[0.9982024,0.0008455189,0.000076335,0.000692333,0.0001467044,0.0000366505],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.000121679,0.001439254,0.2560563,0.0004451305,0.0005210622,0.00001568279,0.01132995,0.0005401376,0.02363232,0.419128,0.0008412039,0.2859293],"study_design_scores_gemma":[0.003445887,0.001014053,0.6797701,0.0001658585,0.0001302288,0.00001499515,0.01133948,0.1728215,0.0828783,0.03773015,0.009709512,0.0009799883],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2457824,0.000249968,0.7485644,0.003841678,0.0002358116,0.0005761366,0.000002175392,0.0001305734,0.0006169035],"genre_scores_gemma":[0.9927224,0.000001624887,0.006583186,0.0003704879,0.00001265326,0.00001827756,6.170137e-7,0.000002687606,0.0002880892],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.74694,"threshold_uncertainty_score":0.30183,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04284217487403143,"score_gpt":0.3160113045522304,"score_spread":0.273169129678199,"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."}}