SN11 debates re-evaluation mechanism, reports scoring bug

Trajectory-RL team discussed whether to re-evaluate top-scoring miners to prevent luck-based wins from dominating the leaderboard. Consensus emerged around re-evaluating top 5 miners, though concerns remain about evaluation queue length and benchmark depth. Separately, tao.com submissions scored zero across all scenarios; root cause under investigation.

• •Re-evaluation of top 5 miners proposed to filter lucky high scores
• •Evaluation throughput bottleneck noted; only ~20 miners evaluated daily
• •Benchmark may be converging too quickly; scenario expansion suggested
• •Tao.com scoring anomaly flagged; potential impact on future evaluations unclear