{"id":"W4385573955","doi":"10.18653/v1/2022.emnlp-main.663","title":"Learning with Rejection for Abstractive Text Summarization","year":2022,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"University of Manchester; Cancer Research UK","keywords":"Automatic summarization; Computer science; Artificial intelligence; Inference; Set (abstract data type); Natural language processing; Training set; Baseline (sea); Paraphrase; Machine learning; Decoding methods; Hallucinating","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":[],"consensus_categories":[],"category_scores_codex":[0.0001469552,0.00003517632,0.00003568644,0.00004246555,0.0002716596,0.00004176675,0.0001335785,0.000008739289,0.00003148447],"category_scores_gemma":[0.00001352775,0.00003192786,0.00001239882,0.000133837,0.000003288912,0.0002293313,0.00007084368,0.00008827851,0.000002258853],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005237713,"about_ca_system_score_gemma":0.00003134492,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004747792,"about_ca_topic_score_gemma":0.00001007253,"domain_scores_codex":[0.9995415,0.00002223209,0.00006048573,0.0001726261,0.0001201696,0.00008297756],"domain_scores_gemma":[0.9997591,0.00004428525,0.00004109744,0.0001066304,0.00003504902,0.00001387278],"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":[0.00003368569,0.00004601199,0.001637778,0.000008615591,0.00001599225,0.000002177578,0.001958621,0.6477528,0.00127695,0.1280704,0.0002739747,0.218923],"study_design_scores_gemma":[0.0002243675,0.0002460661,0.0006191293,0.000001517444,0.000002175962,0.000008948836,0.0004162894,0.9847422,0.0007196419,0.001459877,0.0114791,0.00008074768],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01231477,0.000003732845,0.9807872,0.0003737506,0.0001123602,0.000127327,2.297357e-7,0.0001274893,0.006153211],"genre_scores_gemma":[0.922941,4.679961e-7,0.07358096,0.00008868799,0.00002661575,0.00005758838,0.00000366466,0.00000416815,0.003296876],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9106262,"threshold_uncertainty_score":0.2089414,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01486012973234072,"score_gpt":0.2302687546234288,"score_spread":0.2154086248910881,"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."}}