{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002118567,0.001859627,0.001679591,0.0008629276,0.0005982735,0.001981559,0.003663493,0.002345336,0.01765941],"category_scores_gemma":[0.01260494,0.000713439,0.001066905,0.0005488822,0.0009067468,0.003647121,0.003446541,0.003072432,0.007037856],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001059333,"about_ca_system_score_gemma":0.001836034,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00379167,"about_ca_topic_score_gemma":0.008681766,"domain_scores_codex":[0.9985292,0.0003526701,0.00009970388,0.0004812428,0.0003460716,0.0001911136],"domain_scores_gemma":[0.9963007,0.00189662,0.0001973238,0.001010041,0.0003711536,0.0002241723],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009870842,0.0007128105,0.002265137,0.0007652846,0.0001567434,0.0005376611,0.0005417778,0.2008886,0.03029104,0.02109814,0.05106118,0.6906944],"study_design_scores_gemma":[0.00007826609,0.00009752446,0.0001750769,0.00002435476,0.00003160406,0.0001153967,0.00006796439,0.9667597,0.01123554,0.0142058,0.007174358,0.00003432754],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02557697,0.0007893937,0.8719388,0.0007741472,0.0002723604,0.0002498018,0.000809257,0.09301225,0.006576932],"genre_scores_gemma":[0.4555973,0.0004267587,0.5155815,0.001167935,0.0002193723,0.0004727833,0.00203632,0.007962978,0.01653517],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01765941,"threshold_uncertainty_score":0.05907667,"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."}}