Less Is More: Training-Free Sparse Attention with Global Locality for Efficient Reasoning
Paper
•
2508.07101
•
Published
•
14
•
2
Lijie Yang
drkylj
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
1 day ago
CaveAgent: Transforming LLMs into Stateful Runtime Operators
upvoted
a
paper
4 months ago
SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable
Sparse-Linear Attention
upvoted
a
paper
5 months ago
What Characterizes Effective Reasoning? Revisiting Length, Review, and
Structure of CoT
Organizations
commented
a
paper
6 months ago