{"id":"W4408595536","doi":"10.5705/ss.202023.0202","title":"Addressing Label Noise in Causation Classification via Kernel Embeddings","year":2025,"lang":"en","type":"article","venue":"Statistica Sinica","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Kernel (algebra); Noise (video); Causation; Computer science; Pattern recognition (psychology); Artificial intelligence; Mathematics; Machine learning; Natural language processing; Pure mathematics; Epistemology; Philosophy","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009192096,0.00010324,0.0001642937,0.0001412115,0.0004317797,0.0001339681,0.0002474183,0.0001359412,0.0003547727],"category_scores_gemma":[0.002679045,0.0001168789,0.00002512207,0.0006437544,0.0005648972,0.0002429902,0.00004025544,0.000192715,0.0001768018],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00029969,"about_ca_system_score_gemma":0.0003661083,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004398669,"about_ca_topic_score_gemma":0.008191699,"domain_scores_codex":[0.9983035,0.0002676236,0.0004613423,0.0003199615,0.0003102759,0.0003373385],"domain_scores_gemma":[0.9984508,0.0009425888,0.0001342234,0.0002132144,0.0001816891,0.00007744483],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00003435461,0.0001272701,0.00351768,0.00001800348,0.000009004746,0.00000570348,0.00558827,0.00002086848,0.002300735,0.8289733,0.003652944,0.1557519],"study_design_scores_gemma":[0.0007936963,0.0001166756,0.09876062,0.0006959448,0.0001188477,9.846979e-7,0.01773978,0.07356183,0.002127212,0.6370953,0.1679754,0.00101369],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.05324618,0.0001737983,0.4055398,0.0142985,0.001931881,0.001197245,0.0000801568,0.0003558433,0.5231766],"genre_scores_gemma":[0.9885822,0.00004595393,0.008278516,0.0003328821,0.00008739922,0.00003544044,0.00001736685,0.00001007982,0.002610182],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.935336,"threshold_uncertainty_score":0.6649501,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1676556962956567,"score_gpt":0.4664966625380305,"score_spread":0.2988409662423739,"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."}}