UC Berkeley 新論文提出動態重組編碼AI 助理團隊的方法,在 MLA decode 上達到 FlashInfer 1.21 倍速度,並在 47 項生產工作負載中的 36 項勝出。
AI 伴隨問答
針對這篇報導有疑問?可直接點選快捷問題或向 AI 追問背景脈絡與技術細節:
研究以共享檔案系統累積過往團隊經驗,並在進度停滯時重組角色,呈現長時間 GPU 核心最佳化的一種多AI 助理協作方法。
New UC Berkeley paper shows that a supervisor that keeps reorganizing a team of coding agents can beat expert GPU kernels where single agents stall.
Coding agents make fast early progress on GPU kernels, then settle on easier designs and plateau far below expert code. On the MLA decode kernel, Claude Code stalled at 0.51 times FlashInfer's speed.
Evolving Agent Teams adds a supervisor that diagnoses each plateau and rebuilds the team, inventing roles like a profiler, a debugger, or a knowledge integrator. A shared file system keeps lessons from every past team.
Under the same token budget it reached 0.74 times FlashInfer on MLA decode, against 0.31 for a fixed 4-role team. With more budget it hit 1.21 times FlashInfer's speed, winning on 36 of 47 production workloads.
For long, hard optimization runs, have a supervisor restructure agent roles when progress stalls instead of running 1 fixed agent setup.