{"id":"W4388184389","doi":"10.48550/arxiv.2310.20059","title":"Concept Alignment as a Prerequisite for Value Alignment","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Psychology of Moral and Emotional Judgment","field":"Neuroscience","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Value (mathematics); Artificial intelligence; Representation (politics); Human–computer interaction; Machine learning","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01039801,0.0006781598,0.0007673246,0.00116344,0.001761373,0.003889428,0.001822121,0.002548829,0.007464204],"category_scores_gemma":[0.05706056,0.0009823916,0.001224289,0.0008599964,0.005772398,0.01443355,0.005431784,0.005418192,0.001294969],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001694219,"about_ca_system_score_gemma":0.002740632,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001534895,"about_ca_topic_score_gemma":0.00100676,"domain_scores_codex":[0.9877244,0.004782152,0.0008868603,0.002556213,0.003261993,0.0007883138],"domain_scores_gemma":[0.9690368,0.01493192,0.003434637,0.006519953,0.004369895,0.00170685],"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":[0.0001751212,0.0002120349,0.004606352,0.0002830184,0.00009705388,0.0004744203,0.003245152,0.01424207,0.01120917,0.8831045,0.002113183,0.0802378],"study_design_scores_gemma":[0.00002730022,0.00007471003,0.001539447,0.00005788006,0.00003001055,0.0002428362,0.0004095955,0.05147712,0.005483908,0.9343985,0.006220489,0.00003824733],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04770909,0.0001830447,0.9257434,0.002270594,0.0001043771,0.0001703966,0.00009018749,0.0004742859,0.02325476],"genre_scores_gemma":[0.6921659,0.0001297728,0.3041631,0.0003911208,0.00005750225,0.00020106,0.0002603042,0.0001872826,0.002444045],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01039801,"threshold_uncertainty_score":0.05499059,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.259906587736641,"score_gpt":0.2523106714167854,"score_spread":0.007595916319855567,"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."}}