{"id":"W6893565298","doi":"10.5281/zenodo.3997149","title":"News Corpus for RUSSE'2020 task","year":2020,"lang":"en","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Text Readability and Simplification","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Optech (Canada)","funders":"","keywords":"Task (project management); Corpus linguistics; Key (lock); Headline","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":[0.001048408,0.001677881,0.001065998,0.00374785,0.001390069,0.001210977,0.0009689974,0.001350526,0.05374707],"category_scores_gemma":[0.0029458,0.0005429348,0.0007952353,0.002630639,0.0005186358,0.001040213,0.001517899,0.001285787,0.0501846],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007229077,"about_ca_system_score_gemma":0.001420942,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00737683,"about_ca_topic_score_gemma":0.008936232,"domain_scores_codex":[0.9989422,0.0002849553,0.0001290275,0.0002730064,0.0002529994,0.0001177076],"domain_scores_gemma":[0.9985587,0.000466047,0.00007247949,0.0002761993,0.0005053988,0.0001211827],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0006888115,0.0002209196,0.0009264557,0.002175624,0.00009975437,0.000695029,0.0005081213,0.001113045,0.01783985,0.002715423,0.9184785,0.05453856],"study_design_scores_gemma":[0.0004937913,0.0001751661,0.01379882,0.0002627746,0.0001899084,0.001054198,0.0006523057,0.003750597,0.01282653,0.001296754,0.9654242,0.00007498651],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.03579672,0.001866978,0.01050339,0.0007305237,0.00112993,0.0005783087,0.9079927,0.007359686,0.03404172],"genre_scores_gemma":[0.01686285,0.0003054379,0.006376453,0.000087803,0.0001002755,0.0004885128,0.9662773,0.001090332,0.008411145],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.05374707,"threshold_uncertainty_score":0.1798019,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04795015764437081,"score_gpt":0.2435366356435991,"score_spread":0.1955864779992283,"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."}}