{"id":"W4394782008","doi":"10.1145/3613904.3642934","title":"Unraveling the Dilemma of AI Errors: Exploring the Effectiveness of Human and Machine Explanations for Large Language Models","year":2024,"lang":"en","type":"preprint","venue":"","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada; University of Waterloo","keywords":"Dilemma; Computer science; Artificial intelligence; Natural language processing; Human–machine system; Cognitive science; Machine learning; Epistemology; Psychology; Philosophy","routes":{"ca_aff":true,"ca_fund":true,"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":[0.02172434,0.0007088587,0.0004941341,0.001345108,0.0006357118,0.002354054,0.001020262,0.001464606,0.001898317],"category_scores_gemma":[0.2039872,0.0003961072,0.0004548506,0.0004924923,0.002066184,0.006487157,0.002373365,0.001725363,0.0003127869],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009459209,"about_ca_system_score_gemma":0.0008724183,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001690697,"about_ca_topic_score_gemma":0.00167478,"domain_scores_codex":[0.9883375,0.00877875,0.0004246222,0.00106515,0.001158271,0.0002357064],"domain_scores_gemma":[0.7031314,0.2706851,0.01075562,0.009553197,0.004771915,0.001102802],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003556175,0.0006689249,0.2537837,0.003318386,0.0008220422,0.000930355,0.1014238,0.03087314,0.02889521,0.02929388,0.005557492,0.540877],"study_design_scores_gemma":[0.0006327021,0.002697889,0.1933226,0.001135797,0.0007640925,0.001745857,0.03055045,0.5339146,0.04930259,0.1603711,0.02512667,0.000435734],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9360967,0.001130291,0.05617506,0.001758139,0.00004569346,0.0001236031,0.0002273627,0.0005443791,0.003898788],"genre_scores_gemma":[0.9823577,0.0001173602,0.0169414,0.00009511697,0.00002034177,0.0000349357,0.0001443785,0.00005328053,0.0002353514],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02172434,"threshold_uncertainty_score":0.1148907,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08759386332772177,"score_gpt":0.3359635388024949,"score_spread":0.2483696754747731,"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."}}