weizechen's picture
Create README.md
c443fa4 verified
metadata
datasets:
  - weizechen/RL-Compositionality-Stage1-RFT-Data
base_model:
  - meta-llama/Llama-3.1-8B-Instruct

The model after Stage 1 RFT.

Paper: https://huggingface.co/papers/2509.25123

Code: https://github.com/PRIME-RL/RL-Compositionality