{"id":"W4410629910","doi":"10.1016/j.neucom.2025.130470","title":"SAGE: Self-evolving Agents with Reflective and Memory-augmented Abilities","year":2025,"lang":"en","type":"article","venue":"Neurocomputing","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"Science and Technology Planning Project of Shenzhen Municipality","keywords":"Computer science; SAGE; Artificial intelligence; Cognitive science; Cognitive psychology; Psychology","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":[],"consensus_categories":[],"category_scores_codex":[0.0002385954,0.0001861096,0.0001819093,0.0001631511,0.0003616615,0.0003397541,0.0004688522,0.00004174175,0.000004736064],"category_scores_gemma":[0.00009662323,0.000170844,0.00002909579,0.0005409893,0.00005152328,0.0003366351,0.0005992646,0.0002732135,0.00000495906],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007915647,"about_ca_system_score_gemma":0.00007367175,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001451802,"about_ca_topic_score_gemma":0.000001410869,"domain_scores_codex":[0.9985333,0.0001137003,0.0002391645,0.0005021933,0.0002601643,0.0003514651],"domain_scores_gemma":[0.9990094,0.0003237892,0.0001204319,0.0003926383,0.00009420092,0.00005950145],"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":[0.00004621095,0.0001934325,0.052037,0.000728123,0.0003867804,0.0001768507,0.01635721,0.806729,0.001785288,0.01465836,0.002187831,0.1047139],"study_design_scores_gemma":[0.0005488124,0.0001502497,0.02131821,0.0001682186,0.00001604059,0.00001779628,0.0001451434,0.9758537,0.00070454,0.00017584,0.0007147634,0.0001867058],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07694993,0.00006785586,0.9038557,0.0003885672,0.000283982,0.0003206647,1.30918e-7,0.0005849375,0.01754823],"genre_scores_gemma":[0.9298596,0.000006495829,0.06842996,0.0008193501,0.00003533419,0.000007254879,5.299011e-7,0.00001289971,0.0008285553],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8529097,"threshold_uncertainty_score":0.6966817,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00843311187358675,"score_gpt":0.2504298560117046,"score_spread":0.2419967441381178,"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."}}