Yang Chen (@ychennlp) 's Twitter Profile
Yang Chen

@ychennlp

accelerating @NVIDIA, phd @gtcomputing 🧊locked in

ID: 1043344251780833280

linkhttp://edchengg.github.io calendar_today22-09-2018 03:40:28

84 Tweet

844 Takipçi

491 Takip Edilen

Yang Chen (@ychennlp) 's Twitter Profile Photo

The first thing we did was to make sure the eval setup is correct! We spend a lot of time to make sure our eval can - accurately reproduce the DeepSeek-R1 numbers on AIME, LiveCodeBench - it's IMPOSSIBLE to track the RL progress without a good eval set up (e.g., we see AIME up