Recommended Papers
A curated list of recommended papers for the ML Reading Group.
Machine Learning (ML)
- Recursive Language Models (2025)
- Transformers without Normalization (2025)
- Hardware Scaling Trends and Diminishing Returns in Large-Scale Distributed Training (2025)
- ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution (NeurIPS 2025)
- Flow Matching for Generative Modeling
Reinforcement Learning (RL)
- Partially Observable Reinforcement Learning with Memory Traces (ICML 2025)
- CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning (2025)
- Revisiting the Minimalist Approach to Offline Reinforcement Learning
- Flow Matching Policy Gradients
- Normalizing Flows are Capable Models for RL
- Behavior Regularized Offline Reinforcement Learning
- Training Diffusion Models with Reinforcement Learning