{"id":"W4417455900","doi":"10.26434/chemrxiv-2025-p3wd5","title":"Training Models of Atomic Charge by Predicting a Generalized Force","year":2025,"lang":"","type":"article","venue":"ChemRxiv","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Electronegativity; Charge (physics); Ab initio; Partial charge; Atomic charge; Training (meteorology); Ab initio quantum chemistry methods; Matching (statistics)","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.003384868,0.0005703219,0.001129371,0.0002493531,0.0004818484,0.0003259764,0.001685659,0.0003305731,0.002203288],"category_scores_gemma":[0.0008411867,0.0005856323,0.0002331995,0.0008998959,0.0007014348,0.0006958286,0.000616583,0.000361328,0.0000853231],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001570708,"about_ca_system_score_gemma":0.0005655736,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002372531,"about_ca_topic_score_gemma":0.000003284004,"domain_scores_codex":[0.9948394,0.0004198296,0.001545322,0.001293506,0.0007086355,0.001193249],"domain_scores_gemma":[0.9972668,0.0003143889,0.0009585848,0.001016368,0.0002215023,0.0002224065],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001079868,0.0000907951,0.0005753747,0.0005913732,0.0000354308,0.00000264735,0.002950789,0.003645128,0.9827932,0.00722035,0.001414391,0.0005725329],"study_design_scores_gemma":[0.0009980571,0.00005830146,0.00005410068,0.0005647551,0.00007685339,0.000005230386,0.0002008055,0.3476643,0.6449147,0.004420475,0.0006646291,0.0003778377],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9593372,0.001281003,0.02206776,0.0004629935,0.002487554,0.0006081618,0.00005065833,0.0001967199,0.01350795],"genre_scores_gemma":[0.9845355,0.0001506853,0.006703541,0.0003516274,0.0001719498,0.00006535818,0.00001654638,0.00005323371,0.007951521],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3440191,"threshold_uncertainty_score":0.9996595,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02779927763706789,"score_gpt":0.2850339027637414,"score_spread":0.2572346251266734,"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."}}