{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004600294,0.001470554,0.001325803,0.001145638,0.000523895,0.001652319,0.002065972,0.001603249,0.002186613],"category_scores_gemma":[0.01713512,0.0004342201,0.00102829,0.000778146,0.0008403483,0.002839096,0.00142891,0.002770938,0.001586863],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007405458,"about_ca_system_score_gemma":0.0007615265,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001710872,"about_ca_topic_score_gemma":0.002437908,"domain_scores_codex":[0.9973649,0.001398432,0.0001730409,0.0005285471,0.0004141633,0.0001209807],"domain_scores_gemma":[0.991916,0.005076246,0.0007606552,0.0009266137,0.001118961,0.00020151],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001176691,0.0003593787,0.002484419,0.0006036305,0.0002721053,0.0002256882,0.0008676994,0.2784273,0.02519998,0.01215489,0.0109014,0.6673268],"study_design_scores_gemma":[0.00005455925,0.0002485509,0.0003872834,0.0000296854,0.00005387782,0.00006562482,0.00006866419,0.9779266,0.007873834,0.009739425,0.003523588,0.00002834768],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02705595,0.0009059652,0.9658157,0.0004485054,0.0001034415,0.0001242768,0.0001773786,0.004069987,0.001298723],"genre_scores_gemma":[0.612763,0.0007587303,0.3737474,0.0006916163,0.0004520446,0.000446092,0.0028962,0.000875216,0.007369629],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004600294,"threshold_uncertainty_score":0.02432901,"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."}}