MC-Sparse: Deconstructing and Closing the Dense-Sparse Attention Gap in Diffusion Transformers Paper • 2610.06801 • Published Oct 5 • 37
EnterpriseClawBench: Benchmarking Agents from Real Workplace Sessions Paper • 2606.23654 • Published Jun 22 • 81
Examining Forgetting in Continual Pre-training of Aligned Large Language Models Paper • 2401.03129 • Published Jan 6, 2024 • 1