HySparse: A Hybrid Sparse Attention Architecture with Oracle Token Selection and KV Cache Sharing Paper • 2602.03560 • Published 3 days ago • 40
Less Is More: Training-Free Sparse Attention with Global Locality for Efficient Reasoning Paper • 2508.07101 • Published Aug 9, 2025 • 14
SeerAttention-R: Sparse Attention Adaptation for Long Reasoning Paper • 2506.08889 • Published Jun 10, 2025 • 23