{"id":"W3096831026","doi":"10.18653/v1/2020.coling-main.208","title":"Automatic Detection of Machine Generated Text: A Critical Survey","year":2020,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Compute Canada","keywords":"Computer science; Generative grammar; Data science; Natural language processing; Artificial intelligence; Product (mathematics); Machine learning; Information retrieval","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0004573716,0.000165391,0.0003156807,0.00008919749,0.0000324211,0.0001054847,0.0008119889,0.0001653252,0.00006016358],"category_scores_gemma":[0.0005611936,0.0001520038,0.0000757779,0.000214269,0.00002620185,0.00007894364,0.001286515,0.0003743637,0.00002534],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003849024,"about_ca_system_score_gemma":0.0001392969,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001468314,"about_ca_topic_score_gemma":0.0002608817,"domain_scores_codex":[0.9983479,0.0002569661,0.0004248545,0.0005311758,0.0002781424,0.0001609621],"domain_scores_gemma":[0.9987223,0.0002018983,0.00009166695,0.0007209581,0.0001704147,0.00009279196],"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":[0.00003566673,0.0005741437,0.003031898,0.003815424,0.0004258206,0.0001168094,0.002710079,0.03409409,0.06626854,0.08022215,0.0008834329,0.8078219],"study_design_scores_gemma":[0.0000683587,0.00002481277,0.003605914,0.00003105817,0.000007053476,0.000004479964,0.000001200729,0.9856534,0.00621819,0.004232186,0.000009729656,0.0001436036],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04069993,0.00009710309,0.9569347,0.0007435817,0.0006917479,0.0001441279,0.000009749858,0.0003509074,0.0003281423],"genre_scores_gemma":[0.8367606,0.000003056499,0.1630083,0.0001329326,0.00003677531,0.000009643316,0.000007969993,0.000008543921,0.00003221816],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9515593,"threshold_uncertainty_score":0.6198536,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0582700043796482,"score_gpt":0.3015808177752705,"score_spread":0.2433108133956223,"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."}}