Accelerating Diffusion LLMs via Adaptive Parallel Decoding Paper • 2506.00413 • Published May 31, 2025 • 9
Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning Paper • 2506.01939 • Published Jun 2, 2025 • 187
CoRNStack Collection State-of-the-art code retrieval and re-ranking models and datasets • 9 items • Updated Mar 26, 2025 • 19
view article Article Accelerating LLM Code Generation Through Mask Store Streamlining Jan 17, 2025 • 3