Mela: Test-Time Memory Consolidation based on Transformation Hypothesis Paper • 2605.10537 • Published May 11 • 7
Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention Paper • 2605.22791 • Published May 21 • 31
Running 245 The ultimate guide to RL environments: building and scaling them in the LLM era 📝 245 Building and scaling RL environments for LLM training