{"id":"W4401043372","doi":"10.18653/v1/2024.naacl-short.19","title":"Separately Parameterizing Singleton Detection Improves End-to-end Neural Coreference Resolution","year":2024,"lang":"en","type":"article","venue":"","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"McGill University; Canadian Institute for Advanced Research; Nvidia","keywords":"Singleton; Coreference; Computer science; End-to-end principle; Resolution (logic); Artificial intelligence","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003781473,0.002307407,0.002072206,0.002004899,0.001429899,0.002252078,0.003755216,0.003951813,0.008018545],"category_scores_gemma":[0.01691741,0.0008829008,0.001091398,0.001743561,0.0007022012,0.006408917,0.003880906,0.003484004,0.006970261],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008948237,"about_ca_system_score_gemma":0.001775718,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006043118,"about_ca_topic_score_gemma":0.01597103,"domain_scores_codex":[0.9968881,0.0009361748,0.0002021001,0.001281927,0.0003817408,0.0003098875],"domain_scores_gemma":[0.9946243,0.003040099,0.0001386534,0.0009243685,0.001097513,0.0001751349],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001179261,0.0004814552,0.00170433,0.0002186514,0.0002318538,0.000173469,0.0002124414,0.03179343,0.02654343,0.001584401,0.01879941,0.9170779],"study_design_scores_gemma":[0.00008332914,0.0001966842,0.001697615,0.00005927524,0.0002038956,0.0002587138,0.0001872099,0.9384195,0.04390651,0.009791745,0.005123273,0.0000722403],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1228743,0.005725113,0.8355109,0.001058977,0.001132793,0.0002064701,0.00102984,0.02106458,0.011397],"genre_scores_gemma":[0.596639,0.001280534,0.3777378,0.0009877811,0.0003727134,0.0002434012,0.004260816,0.002385444,0.01609252],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008018545,"threshold_uncertainty_score":0.02682471,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03356020446073785,"score_gpt":0.2770017379538318,"score_spread":0.2434415334930939,"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."}}