{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003625889,0.0003562507,0.0003843553,0.0002136226,0.0002399887,0.0006131738,0.000876013,0.0007420813,0.004760497],"category_scores_gemma":[0.001972366,0.000194325,0.0003245534,0.0001600242,0.0006644951,0.0007435668,0.00147358,0.0008447961,0.001114939],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001769458,"about_ca_system_score_gemma":0.0004089836,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004693866,"about_ca_topic_score_gemma":0.0007139961,"domain_scores_codex":[0.9998645,0.00003692216,0.000009128196,0.00002726921,0.00004735951,0.00001482205],"domain_scores_gemma":[0.9994915,0.0002372865,0.00003887614,0.00009485735,0.00005908927,0.00007840354],"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":[0.0005143791,0.000302132,0.002750844,0.0003279299,0.0001364329,0.0005722535,0.0004616848,0.6335627,0.0384897,0.131339,0.0116511,0.1798919],"study_design_scores_gemma":[0.0001334955,0.0001666119,0.0002939032,0.00001940135,0.00002810852,0.0001593365,0.0000413747,0.9243333,0.008887995,0.0548197,0.01109892,0.00001785879],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09539163,0.0002380225,0.8797206,0.0004087104,0.0002371549,0.0001325256,0.0002046847,0.007777764,0.01588878],"genre_scores_gemma":[0.7161959,0.0002150341,0.2632798,0.0001810831,0.00004740804,0.0003182453,0.0003545338,0.0004391875,0.01896896],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004760497,"threshold_uncertainty_score":0.01592547,"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."}}