MassAlloc Attention: Let Attention Allocate Its Own Compute Paper • 2609.32712 • Published 4 days ago • 57
CoWindow Attention: Full Causal Coverage Is a Collective Property Paper • 2609.32704 • Published 4 days ago • 17
view article Article Trainable Dynamic Mask Sparse Attention: Bridging Efficiency and Effectiveness in Long-Context Language Models wubingheng • Aug 5, 2025 • 7
view article Article Trainable Dynamic Mask Sparse Attention: Bridging Efficiency and Effectiveness in Long-Context Language Models wubingheng • Aug 5, 2025 • 7
🧐Small-Papers Collection Technical support for the SmallDoges series models. • 2 items • Updated Aug 5, 2025 • 2
Concise Reasoning, Big Gains: Pruning Long Reasoning Trace with Difficulty-Aware Prompting Paper • 2505.19716 • Published May 26, 2025 • 4