{"id":"W7082258602","doi":"10.48448/nxyz-de54","title":"Retrieval-Augmented Process Reward Model for Generalizable Mathematical Reasoning","year":2025,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Geochemistry and Geologic Mapping","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Generalization; Consistency (knowledge bases); Process (computing); Deductive reasoning; Model-based reasoning; Opportunistic reasoning","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":[],"consensus_categories":[],"category_scores_codex":[0.004609907,0.001134978,0.001193167,0.001252441,0.0006436004,0.002128423,0.003487356,0.00195975,0.007666753],"category_scores_gemma":[0.02195328,0.0005938624,0.001954977,0.001175036,0.001704237,0.005660268,0.002609156,0.003951164,0.001853611],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002530885,"about_ca_system_score_gemma":0.002796433,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007155116,"about_ca_topic_score_gemma":0.01073163,"domain_scores_codex":[0.9974211,0.001022733,0.0001626086,0.0007789669,0.000430406,0.0001840417],"domain_scores_gemma":[0.9927849,0.004374547,0.0006412,0.001210012,0.0007220689,0.0002673126],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0004512453,0.0003691748,0.003542434,0.0004359997,0.0001387342,0.0002065435,0.0003727714,0.6683412,0.003401553,0.1441973,0.01105809,0.167485],"study_design_scores_gemma":[0.00002617243,0.00004015658,0.0001889524,0.00001266571,0.00001117074,0.00002252519,0.00001051967,0.931526,0.0004872399,0.06627056,0.001392819,0.00001122325],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02310494,0.0006143707,0.9655565,0.00149711,0.00007074091,0.0001900558,0.001108182,0.003628026,0.004230103],"genre_scores_gemma":[0.5536705,0.0005319619,0.4334693,0.0007781547,0.0001302469,0.0007572672,0.002821538,0.0005259128,0.007315136],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007666753,"threshold_uncertainty_score":0.02564788,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02482961139892289,"score_gpt":0.2977676714290438,"score_spread":0.2729380600301209,"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."}}