{"id":"W4214826913","doi":"10.1109/taslp.2022.3155281","title":"Dealing With Hierarchical Types and Label Noise in Fine-Grained Entity Typing","year":2022,"lang":"en","type":"article","venue":"IEEE/ACM Transactions on Audio Speech and Language Processing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"Fundamental Research Funds for the Central Universities; State Key Laboratory of Software Development Environment; National Natural Science Foundation of China","keywords":"Benchmarking; Computer science; Noise (video); Hierarchical organization; Hierarchical database model; Artificial intelligence; Data mining; Pattern recognition (psychology); Machine learning; Image (mathematics)","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":[0.0003362322,0.0001973446,0.0002107612,0.0003386346,0.0005476828,0.0002322605,0.0004166111,0.00005837737,0.00002648633],"category_scores_gemma":[0.00003345201,0.0001755279,0.00002443183,0.0006427672,0.0000793621,0.0004317953,0.00004852296,0.0006654036,0.000001000107],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006665112,"about_ca_system_score_gemma":0.00008645318,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001272662,"about_ca_topic_score_gemma":0.000242466,"domain_scores_codex":[0.9986608,0.00007090783,0.0002022705,0.0004972361,0.0002598916,0.0003088624],"domain_scores_gemma":[0.9993942,0.00008497627,0.00008490611,0.0002981213,0.00004185886,0.00009590005],"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":[0.0001004945,0.0001274892,0.0002092471,0.0001414423,0.00001542442,0.0002017586,0.003688843,0.0001507339,0.01155642,0.0002648892,0.000005753821,0.9835375],"study_design_scores_gemma":[0.01811323,0.00447791,0.002984423,0.004307143,0.0004447163,0.007439033,0.007673713,0.4407574,0.4472713,0.05639893,0.002307412,0.007824857],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2692484,0.009002364,0.7191162,0.001534451,0.00008686413,0.0002832774,0.00001270101,0.0005668869,0.0001488198],"genre_scores_gemma":[0.7185103,0.0000248114,0.2810329,0.0002565699,0.0000189498,0.00003131611,0.000001983025,0.00001621104,0.0001069483],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9757127,"threshold_uncertainty_score":0.7157818,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01122754845453324,"score_gpt":0.2616009364783938,"score_spread":0.2503733880238606,"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."}}