Alert button

Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time

Oct 26, 2023
Zichang Liu, Jue Wang, Tri Dao, Tianyi Zhou, Binhang Yuan, Zhao Song, Anshumali Shrivastava, Ce Zhang, Yuandong Tian, Christopher Re, Beidi Chen

Figure 1 for Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time
Figure 2 for Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time
Figure 3 for Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time
Figure 4 for Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: