{"id":"W7000228838","doi":"","title":"Exploring Sentence Variations with Bilingual Corpora","year":2005,"lang":"en","type":"article","venue":"NPARC","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Sentence; Grammar; Context (archaeology); Machine translation; Parallel corpora; Translation (biology)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002511535,0.00007422805,0.00007102809,0.00005668419,0.0001482395,0.0000757965,0.0003103479,0.00002076253,0.00005827632],"category_scores_gemma":[0.00004033823,0.00006229095,0.00002121885,0.0003554261,0.00002371617,0.0006701518,0.00007427023,0.0001182468,0.0001433589],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003296146,"about_ca_system_score_gemma":0.0001064838,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000381079,"about_ca_topic_score_gemma":0.000007320456,"domain_scores_codex":[0.9992337,0.00003604637,0.000121357,0.0002191027,0.0001967324,0.0001930524],"domain_scores_gemma":[0.9994448,0.00005850516,0.00005464349,0.0002803431,0.00007848848,0.00008315577],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000008123156,0.00005310938,0.002392873,0.000005790043,0.00001152291,0.00001484431,0.001947996,0.0007175194,0.004686832,0.9094939,0.0001668879,0.08050065],"study_design_scores_gemma":[0.001875564,0.0004069459,0.01597969,0.0001649144,0.00003527557,0.0002568783,0.0004740586,0.7469188,0.1163589,0.05872042,0.05738526,0.001423283],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04996264,0.000009203596,0.9412341,0.003064695,0.0001665634,0.00006939763,0.000001401896,0.000232687,0.005259327],"genre_scores_gemma":[0.7787013,0.000004301509,0.2207575,0.0002246953,0.0001031961,0.00001019473,0.000001604198,0.000003816307,0.0001934306],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8507735,"threshold_uncertainty_score":0.2540151,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1264075600395029,"score_gpt":0.2753025695211809,"score_spread":0.148895009481678,"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."}}