Notes
2026
2026-10-17 — 从 RLVR 到 RLCD:当强化学习不再只奖励“正确”
2026-10-06 — I got this:拥有不用 AI 的权利
2026-09-09 — 数学从来不只是答案
2026-09-02 — 不需要想的思考 - Recurrent Depth
2026-09-02 — Thinking Without Talking - Recurrent Depth
2026-08-29 — Network richness and feature learning
2026-08-29 — A note about pictures on this website
2026-08-21 — Reading notes on lazy and rich regimes
2026-08-20 — 广东与重庆
2026-08-18 — Sites and charts
2026-08-15 — AI
2026-08-13 — A numerical method I keep returning to
2026-08-06 — Random thoughts on research and life
2026-07-28 — Visa limbo, time zones, and coffee