Dr. GRPO: Understanding R1-Zero-Like Training with Zichen Liu
Deep Learning with Yacine · 68:33 · Watch on YouTube
The interview argues that much of the R1-Zero "magic" is overstated. Its first author is rendered as "Zean Wu" in the transcript (Sea AI Lab / NUS). The "aha moment" of self-reflection already exists in the DeepSeek-V...