SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuning
Paper
•
2602.13515
•
Published
•
33
umip
umip
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
about 9 hours ago
SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuning
upvoted
a
paper
1 day ago
Unified Latents (UL): How to train your latents
upvoted
a
paper
3 days ago
Sanity Checks for Sparse Autoencoders: Do SAEs Beat Random Baselines?
Organizations
None yet