{"id":"W4408352598","doi":"10.1109/icassp49660.2025.10889047","title":"Learning from Ambiguous Data with Hard Labels","year":2025,"lang":"en","type":"article","venue":"","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Artificial intelligence","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":[0.009801585,0.001767967,0.002100606,0.001658314,0.001237434,0.003254964,0.003779262,0.003092509,0.001671606],"category_scores_gemma":[0.03499477,0.0008816372,0.001354945,0.001583566,0.004366456,0.006230043,0.005737943,0.006725576,0.0008276015],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001717436,"about_ca_system_score_gemma":0.001547439,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001694058,"about_ca_topic_score_gemma":0.002021035,"domain_scores_codex":[0.9927665,0.002813576,0.0003739498,0.002291403,0.001411165,0.0003433084],"domain_scores_gemma":[0.9789123,0.01245555,0.00169582,0.005092383,0.001290521,0.000553408],"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":[0.0007598706,0.0005471045,0.0199299,0.0006239396,0.0002542065,0.0006046891,0.001358373,0.4724608,0.00818827,0.1202496,0.01260711,0.3624161],"study_design_scores_gemma":[0.00004176594,0.00007695753,0.001016535,0.00007236527,0.00002431328,0.0001259709,0.0001396007,0.8364441,0.002771618,0.1569689,0.00227683,0.00004105599],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03039006,0.0003371352,0.9663695,0.000717442,0.000069223,0.00007470472,0.0002348297,0.0006585715,0.001148542],"genre_scores_gemma":[0.6653516,0.0003412802,0.328237,0.00133663,0.0002588625,0.0003810154,0.002028232,0.0002661443,0.001799223],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009801585,"threshold_uncertainty_score":0.05183631,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03455206419651542,"score_gpt":0.2792409356508822,"score_spread":0.2446888714543668,"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."}}