Dr. GRPO

A technical AI safety researcher argues that safety-focused researchers should engage with capabilities research, highlighting On-Policy Self-Distillation (OPSD) as a promising alternative to GRPO for…

A developer wrote a fused decode-attention kernel that ran 2.2× faster than the baseline in microbenchmarks, but when integrated into a HuggingFace `generate` call for an RL training loop, the decode …