LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches Paper • 2610.06647 • Published 3 days ago • 107
LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches Paper • 2610.06647 • Published 3 days ago • 107
OSWorld-Pro: Process-based Evaluation for Computer Use Agents Paper • 2609.24890 • Published 17 days ago • 24
OSWorld-Pro: Process-based Evaluation for Computer Use Agents Paper • 2609.24890 • Published 17 days ago • 24
Agora: Git as Shared Memory for Collective AutoResearch Paper • 2609.18094 • Published 22 days ago • 58
Agora: Git as Shared Memory for Collective AutoResearch Paper • 2609.18094 • Published 22 days ago • 58
Residual Stream Duality in Modern Transformer Architectures Paper • 2603.16039 • Published Mar 17 • 4
Residual Stream Duality in Modern Transformer Architectures Paper • 2603.16039 • Published Mar 17 • 4