{"id":"W6976785268","doi":"10.60692/xekv8-pty78","title":"Datasets for \"Discourse-Aware Unsupervised Summarization for Long Scientific Documents\"","year":2021,"lang":"en","type":"article","venue":"Greater South Information System","topic":"Climate Change and Health Impacts","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Automatic summarization; Key (lock); Feature (linguistics); Unsupervised learning; Pattern recognition (psychology)","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":[0.002719285,0.001723182,0.0008002231,0.004342339,0.002242076,0.001408225,0.002309486,0.002437937,0.007865535],"category_scores_gemma":[0.01191633,0.0004662674,0.00157945,0.00408959,0.0007164109,0.001647785,0.002416481,0.00207566,0.006052109],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001751307,"about_ca_system_score_gemma":0.003747702,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01545764,"about_ca_topic_score_gemma":0.03476702,"domain_scores_codex":[0.9974244,0.000772511,0.0003799585,0.0005694808,0.0006660115,0.0001877234],"domain_scores_gemma":[0.9927766,0.002973051,0.0006059812,0.001148702,0.002021633,0.000474065],"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":[0.00115293,0.001268143,0.004924963,0.006768406,0.0004618001,0.0006108208,0.001617495,0.009702864,0.02387025,0.005055639,0.8109774,0.1335894],"study_design_scores_gemma":[0.001329257,0.0005016515,0.03471433,0.0007461453,0.0003880531,0.0006075996,0.002395851,0.04566198,0.04341182,0.007628587,0.862285,0.0003296997],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0366343,0.001814706,0.01935583,0.00186599,0.0004060955,0.001532135,0.9183595,0.01359476,0.006436682],"genre_scores_gemma":[0.01309327,0.0002586005,0.04184142,0.0001361327,0.00005429283,0.001517895,0.9408046,0.0002255198,0.0020682],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01545764,"threshold_uncertainty_score":0.03073531,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07499105048947406,"score_gpt":0.296938113143088,"score_spread":0.2219470626536139,"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."}}