{"id":"W6939282806","doi":"10.60692/7v46n-wbe56","title":"Datasets for \"Discourse-Aware Unsupervised Summarization for Long Scientific Documents\"","year":2021,"lang":"en","type":"article","venue":"Greater South Information System","topic":"Topic Modeling","field":"Computer 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":[0.002187606,0.001623838,0.0007512881,0.004018606,0.002113196,0.001198438,0.00231474,0.002186168,0.006311475],"category_scores_gemma":[0.008419218,0.0004332028,0.001450747,0.003956882,0.0005770714,0.001394536,0.001946667,0.001887642,0.005468857],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001570318,"about_ca_system_score_gemma":0.003069698,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01222322,"about_ca_topic_score_gemma":0.03250718,"domain_scores_codex":[0.998055,0.0005754615,0.0002584795,0.0004417677,0.0005089227,0.0001603984],"domain_scores_gemma":[0.99439,0.001986605,0.0005236596,0.0009256264,0.00171476,0.0004592573],"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":[0.001228975,0.001499505,0.005531399,0.005494871,0.0004257243,0.0005441614,0.001642498,0.008452454,0.02684189,0.004516584,0.8138082,0.1300137],"study_design_scores_gemma":[0.001463442,0.0006527675,0.05010258,0.0006114038,0.0004274187,0.0006972034,0.002373042,0.04346092,0.04589268,0.006203167,0.8478026,0.0003127039],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.05747366,0.002096027,0.02091319,0.001751679,0.0004320582,0.001600352,0.8938043,0.0140758,0.007852918],"genre_scores_gemma":[0.01479568,0.0002740736,0.03752301,0.0001184958,0.00006373675,0.001364394,0.9430596,0.0002138595,0.002587202],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01222322,"threshold_uncertainty_score":0.02430415,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04695024298544776,"score_gpt":0.2603624005255004,"score_spread":0.2134121575400527,"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."}}