LangChain 发布 Tuned Evaluators，自动评估 Agent 行为，成本降 82%，比前沿模型更准。
LangChain 发布了 LangSmith Tuned Evaluators，能自动为生产环境中的 Agent 行为打分。该功能首发包含 Perceived Error 指标，用于识别 Agent 是否给用户提供了有效帮助。在内部基准测试中，其专用模型表现优于所有测试的前沿模型。该工具将评估成本降低了 82%。
Tuned Evaluators automatically add quality feedback to production traces. They come with the tuned ...
Tuned Evaluators automatically add quality feedback to production traces. They come with the tuned model, prompt, and managed infrastructure already set up, so you can start finding agent behavior that needs attention in a few clicks. Your browser does not support the video tag. 🔗 View on Twitter LangChain @LangChain Introducing LangSmith Tuned Evaluators They automatically score agent behavior in production, starting with Perceived Error. Perceived Error is one of the clearest signals that your agent is giving users a helpful experience. In our benchmark, our specialized model outperformed every frontier model we tested and reduced evaluation cost by 82%. Learn more: langchain.com/blog/introduci… 🔗 View Quoted Tweet 💬 2 🔄 0 ❤️ 12 👀 1860 📊 3 ⚡