AlphaGo Moment for Model Architecture Discovery.

Reproducible and flexible LLM evaluations for scientific reasoning.

MegaScience: Pushing the Frontiers of Post-Training Datasets for Science Reasoning

Revisiting Mid-training in the Era of Reinforcement Learning Scaling

[COLM 2025] LIMO: Less is More for Reasoning

[ICML 2025] Programming Every Example: Lifting Pre-training Data Quality Like Experts at Scale

Anole: An Open, Autoregressive and Native Multimodal Models for Interleaved Image-Text Generation

Efficient Agent Training for Computer Use