https://zhanyuan.dev/Recent content on zhanyuan.devHugoenhttps://zhanyuan.dev/about/Mon, 01 Jan 0001 00:00:00 +0000https://zhanyuan.dev/about/<p>I work on training large language models (SFT, GRPO), building agents, and making tool calling reliable.</p> <p><strong>Email:</strong> <a href="mailto:li949777411@gmail.com">li949777411@gmail.com</a></p>