MARS: Multi-Specialist LLM Relay System for Competitive Programming Paper • 2608.23918 • Published 10 days ago • 9
Running 31 Weight-Space Geometry of Offline Reasoning Training 🧭 31 Interactive weight-space geometry of six reasoning losses
GradMem: Learning to Write Context into Memory with Test-Time Gradient Descent Paper • 2603.13875 • Published Mar 14 • 36
Running on CPU Upgrade 275 The Synthetic Data Playbook: Generating Trillions of the Finest Tokens 📝 275 Visualize synthetic‑data experiments as an interactive bookshelf
SWE-rebench V2: Language-Agnostic SWE Task Collection at Scale Paper • 2602.23866 • Published Feb 27 • 92