{"id":"W4417322872","doi":"10.48550/arxiv.2510.21059","title":"Dynamic Retriever for In-Context Knowledge Editing via Policy Optimization","year":2025,"lang":"","type":"preprint","venue":"ArXiv.org","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Scalability; Task (project management); Software; Labrador Retriever; Source code; Collaborative editing","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001288106,0.0008560125,0.0009379663,0.001259645,0.0007181323,0.0003218276,0.003029398,0.0008431639,0.0000685301],"category_scores_gemma":[0.002210842,0.001044912,0.0004322192,0.002738058,0.0001855557,0.0006481676,0.003106856,0.001944699,0.0002170477],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00134944,"about_ca_system_score_gemma":0.001676192,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002822574,"about_ca_topic_score_gemma":0.0004929106,"domain_scores_codex":[0.9940984,0.0003924762,0.001589322,0.002471019,0.0003466942,0.001102053],"domain_scores_gemma":[0.9944015,0.001110479,0.001017118,0.002345692,0.0008646135,0.0002605856],"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":[0.00005813825,0.0006365799,0.11499,0.000938868,0.000147857,0.000003685511,0.006151462,0.7000315,0.0002493353,0.009340867,0.000157537,0.1672942],"study_design_scores_gemma":[0.001038677,0.00006231516,0.09670562,0.0006490991,0.00005046837,0.000004189449,0.00005756861,0.896889,0.0001146939,0.001236841,0.002487344,0.0007042126],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06442338,0.0004532779,0.9196916,0.007781101,0.002152518,0.002735679,0.00007106497,0.0003170456,0.002374304],"genre_scores_gemma":[0.9141665,0.000125476,0.07997604,0.0005356649,0.0008507497,0.0009839794,0.0001963219,0.00007300497,0.003092289],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8497431,"threshold_uncertainty_score":0.9992001,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02639491692215508,"score_gpt":0.3334317439378213,"score_spread":0.3070368270156662,"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."}}