SLA2: Sparse-Linear Attention with Learnable Routing and QAT
Paper
• 2602.12675 • Published
• 53
Zhennan Wan
zhennan1
AI & ML interests
Natural Language Processing, AI Agents
Recent Activity
upvoted a paper 15 days ago
SLA2: Sparse-Linear Attention with Learnable Routing and QAT upvoted a paper 2 months ago
TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times upvoted a paper 5 months ago
SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable
Sparse-Linear Attention Organizations
