PyTorchCon 2026 演讲：如何把 Agent 轨迹变成模型训练信号
PyTorch 官方转发的一场演讲，讲怎么把 Agent 的评估结果直接变成训练奖励，做 Agent 开发的话值得听听思路。
Agentic AI Foundation 执行总监 Mazin Gilbert 在 PyTorchCon 2026 演讲中把 agentic 栈分为五层：硬件基础设施、软件基础设施、智能、agentic 平台和用户应用，PyTorch 位于智能层，Agentic AI Foundation 位于平台层。他提出双向设计：模型支撑 Agent，同时把 Agent 轨迹、评估、环境和奖励反向喂给模型训练。他描述的开放研究方向是把 Agent 评估转化为 PyTorch 训练中的奖励，并计入安全、任务完成度、精度、效率和能耗五个维度。
How can LLMs learn from what agents do? At #PyTorchCon 2026, Mazin Gilbert, Executive Director of @AgenticAIFdn, examines the relationship between models and agents.
Mazin maps five layers of the agentic stack: hardware infrastructure, software infrastructure, intelligence, the agentic platform, and user applications. He places PyTorch at the intelligence layer and Agentic AI Foundation at the agentic platform layer.
He proposes designing the stack in both directions: models support agents, while agent traces, evaluations, environments, and rewards feed back into how models are built.
As an open research direction, Mazin describes turning agent evaluations into rewards for training in PyTorch, accounting for safety, task completion, precision, efficiency, and energy use.
Watch the keynote livestream: https://t.co/6TtaEkvsHz
