{"id":"W3173327529","doi":"10.1145/3462757.3466149","title":"Lex Rosetta","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"European Regional Development Fund; Masarykova Univerzita; Université de Montréal","keywords":"Annotation; Computer science; Sentence; Pooling; Robustness (evolution); Natural language processing; Artificial intelligence; Context (archaeology); Segmentation; Language model; Geography","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0006111705,0.0001242563,0.0001905934,0.00003614669,0.0003331994,0.0004300854,0.0005899136,0.0003948171,0.01162408],"category_scores_gemma":[0.0004717471,0.0001295566,0.000146768,0.0001365715,0.0004102956,0.0001011901,0.000594556,0.0004254859,0.000505586],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001425391,"about_ca_system_score_gemma":0.0007564233,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.02779833,"about_ca_topic_score_gemma":0.03983188,"domain_scores_codex":[0.9984075,0.0001957114,0.0002336483,0.0003873439,0.0004605087,0.0003152593],"domain_scores_gemma":[0.999063,0.0001161349,0.00007990544,0.0004124118,0.0002059669,0.000122654],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00000446104,0.0001345683,0.001674489,0.00003975124,0.00006718066,0.00006398113,0.04078296,0.00053954,0.0001679425,0.8671297,0.03764593,0.05174947],"study_design_scores_gemma":[0.0000203398,0.00001127283,0.0002314203,0.0001263492,0.00003674355,6.074765e-7,0.03990152,0.000354072,0.006024024,0.1036691,0.8488525,0.0007721143],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.03502374,0.0002530873,0.004873754,0.005692274,0.003483414,0.0002752515,0.000005308837,0.0002922144,0.950101],"genre_scores_gemma":[0.9513987,0.0002327204,0.003109782,0.0006958019,0.001102693,0.00002559851,0.0000209189,0.00001725925,0.04339647],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.916375,"threshold_uncertainty_score":0.9892794,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1335491857873395,"score_gpt":0.4338240010358737,"score_spread":0.3002748152485342,"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."}}