Discovering the Gems in Early Layers: Accelerating Long-Context LLMs with 1000x Input Token Reduction
Paper
•
2409.17422
•
Published
•
25
•
5
Zhenmei Shi
Zhenmei
AI & ML interests
None yet
Recent Activity
upvoted
a
collection
2 days ago
DeepSeek-V3
liked
a Space
4 days ago
mteb/leaderboard_2_demo
liked
a model
9 days ago
voyageai/voyage-multimodal-3
Organizations
Zhenmei's activity
commented
a
paper
4 months ago