{"id":"W4297785970","doi":"10.36227/techrxiv.21067438.v1","title":"A Trustworthy View on XAI Method Evaluation","year":2022,"lang":"en","type":"preprint","venue":"","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"Consistency (knowledge bases); Trustworthiness; Computer science; Centroid; Process (computing); Cluster analysis; Data mining; Order (exchange); Feature (linguistics); Artificial intelligence; Business; Computer security","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":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.2042717,0.002024904,0.002379263,0.006263803,0.003122963,0.01405775,0.005125513,0.004481061,0.002587753],"category_scores_gemma":[0.5044934,0.001742051,0.002005776,0.002758113,0.009199972,0.01238719,0.00920773,0.009278606,0.00106544],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004685695,"about_ca_system_score_gemma":0.006954773,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00271485,"about_ca_topic_score_gemma":0.001980348,"domain_scores_codex":[0.7105219,0.2017833,0.01276362,0.01072781,0.06181367,0.002389678],"domain_scores_gemma":[0.4371603,0.3639218,0.02860477,0.094152,0.07251599,0.003645101],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00121474,0.0006575409,0.0396376,0.002120341,0.001065948,0.0006207639,0.01225812,0.07408869,0.01403058,0.3633367,0.01073046,0.4802385],"study_design_scores_gemma":[0.0002902313,0.0008540653,0.008412998,0.001910023,0.0002456552,0.000718781,0.002001516,0.6004452,0.03093874,0.3264423,0.02744241,0.0002980313],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01840941,0.001121333,0.9707333,0.002867521,0.0001412775,0.0003805145,0.0000741012,0.00109132,0.005181246],"genre_scores_gemma":[0.2303049,0.0004373115,0.7652324,0.0008237966,0.0001614263,0.0007336012,0.0002201353,0.0009044557,0.00118197],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7957283,"threshold_uncertainty_score":0.9812747,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1202530959946061,"score_gpt":0.4160330938312373,"score_spread":0.2957799978366311,"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."}}