{"id":"W4408518028","doi":"10.2139/ssrn.5182425","title":"Sage: Self-Evolving Agents with Reflective and Memory-Augmented Abilities","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Psychology; Cognitive psychology; SAGE; Computer science; Cognitive science; Physics","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","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.001795328,0.0004648862,0.0004668967,0.0003754346,0.0004545281,0.0007225663,0.001449753,0.000228317,0.00001558024],"category_scores_gemma":[0.0001194295,0.0004033778,0.0001267083,0.0003266611,0.00008749213,0.000469833,0.001414195,0.005751992,0.000005380311],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002818191,"about_ca_system_score_gemma":0.005913204,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007735098,"about_ca_topic_score_gemma":0.0001126107,"domain_scores_codex":[0.9955633,0.0002871233,0.0004926836,0.0007135098,0.0006922319,0.002251149],"domain_scores_gemma":[0.9981412,0.0001604984,0.0005427392,0.000729967,0.00029585,0.0001297763],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000251595,0.0003910598,0.008206787,0.001145219,0.008365187,0.0001397484,0.02111485,0.7034021,0.00007134712,0.1869186,0.001191589,0.06880189],"study_design_scores_gemma":[0.003764285,0.002511447,0.002912025,0.002251365,0.0006052136,0.001404264,0.004596701,0.7178751,0.000262527,0.2601281,0.001433309,0.002255673],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007290688,0.003189578,0.9807397,0.0007279926,0.0006204387,0.0004764394,0.000001855533,0.0002721882,0.006681173],"genre_scores_gemma":[0.9227465,0.01046378,0.04361577,0.0003947054,0.0003791486,0.000057646,0.00001093634,0.00006626908,0.02226525],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9371239,"threshold_uncertainty_score":0.9998418,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009187574990174665,"score_gpt":0.2573514460157894,"score_spread":0.2481638710256147,"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."}}