{"id":"W4285171765","doi":"10.18653/v1/2022.acl-long.360","title":"PRIMERA: Pyramid-based Masked Sentence Pre-training for Multi-document Summarization","year":2022,"lang":"en","type":"article","venue":"Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","topic":"Topic Modeling","field":"Computer Science","cited_by":119,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Automatic summarization; Computer science; Encoder; Transformer; Focus (optics); Artificial intelligence; Pyramid (geometry); Sentence; Multi-document summarization; Training set; Natural language processing; Information retrieval; 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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.001946254,0.0001920311,0.0002858427,0.0001025867,0.0009051673,0.00009883079,0.001391463,0.00007050724,0.000001421488],"category_scores_gemma":[0.01016322,0.0001703522,0.0002946767,0.0004527371,0.00005655903,0.0001051806,0.0005636716,0.0002114918,2.683721e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004562008,"about_ca_system_score_gemma":0.0003002155,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003300077,"about_ca_topic_score_gemma":0.000003366459,"domain_scores_codex":[0.9973783,0.00004425678,0.0007309707,0.0004203794,0.001074735,0.0003513039],"domain_scores_gemma":[0.9946828,0.0007170424,0.001835914,0.0001838524,0.002529012,0.00005131462],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00008235298,0.0001699618,0.064684,0.0003060024,0.0001405458,7.270882e-8,0.003942243,0.8662972,0.0009065494,0.06204662,0.0005476542,0.0008767265],"study_design_scores_gemma":[0.001069829,0.0001128468,0.01243103,0.0001227558,0.00008560521,8.006776e-7,0.0004278134,0.9732101,0.001484016,0.008459907,0.002373539,0.0002217782],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1910003,0.0001841646,0.7786371,0.008173574,0.01014287,0.007705552,0.001153455,0.0005434411,0.002459588],"genre_scores_gemma":[0.859892,4.331648e-7,0.1387345,0.0002786061,0.0001844276,0.0001185926,0.00002312206,0.00002355233,0.0007447971],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6688917,"threshold_uncertainty_score":0.9981746,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01958442892774115,"score_gpt":0.2617687011148682,"score_spread":0.242184272187127,"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."}}