Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Agent Continuous Improvement on CoSQL StreamBench (32 held-out episodes)

73.5Pass@4

Memory w/ MEMOPILOT

65.767.72569.7571.775Jun 7, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
73.5
2026.06
70
2026.06
69.5
2026.06
67.5
2026.06
66