{"id":"W3098136301","doi":"10.18653/v1/2020.emnlp-main.748","title":"On Extractive and Abstractive Neural Document Summarization with Transformer Language Models","year":2020,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":187,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; Université de Montréal; Polytechnique Montréal","funders":"","keywords":"Automatic summarization; Computer science; Transformer; Language model; Natural language processing; Artificial intelligence; Fluency; Question answering; Machine learning; Artificial neural network; Information retrieval; Linguistics","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":[0.001477959,0.001338686,0.0007477597,0.001851397,0.0004062157,0.001407389,0.001168633,0.0008829529,0.002488843],"category_scores_gemma":[0.006973643,0.0003691692,0.0008635803,0.001596956,0.0005390279,0.003218131,0.001128982,0.001361675,0.001893906],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008537527,"about_ca_system_score_gemma":0.001157,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004091031,"about_ca_topic_score_gemma":0.008488297,"domain_scores_codex":[0.9991933,0.0002580567,0.00008476919,0.000177992,0.0002330683,0.00005282785],"domain_scores_gemma":[0.9969439,0.001638773,0.0002953148,0.0004114235,0.0006089617,0.0001016335],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004263834,0.0002104726,0.001121353,0.0005515338,0.0001866869,0.0002137194,0.0005009706,0.1216507,0.03937862,0.01264329,0.008895063,0.8142211],"study_design_scores_gemma":[0.00006369646,0.0004001852,0.0006893909,0.00005228267,0.0001280959,0.0001559883,0.0001621318,0.9361052,0.02941521,0.02258538,0.01018753,0.0000548991],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02448143,0.001849439,0.9626402,0.0004921124,0.0001169188,0.0001515205,0.0006733978,0.006778708,0.002816317],"genre_scores_gemma":[0.3460147,0.002004685,0.6344753,0.0004555669,0.0003758164,0.0003724497,0.004754695,0.0007113586,0.01083546],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004091031,"threshold_uncertainty_score":0.008326054,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01696996314490283,"score_gpt":0.2343981941818167,"score_spread":0.2174282310369139,"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."}}