3 citations · 3 across the 5 of their papers we have counts for
3 papers · 1 filter
StagFormer: Time Staggering Transformer Decoding for RunningLayers In Parallel
Dylan Cutler, Arun Kandoor, Nishanth Dikkala +3
Decoding in a Transformer based language model is inherently sequential as a token's embedding needs to pass through all the layers in the network before the generation of the next…
The Power of External Memory in Increasing Predictive Model Capacity
Cenk Baykal, Dylan J Cutler, Nishanth Dikkala +3
One way of introducing sparsity into deep networks is by attaching an external table of parameters that is sparsely looked up at different layers of the network. By storing the bul…
Alternating Updates for Efficient Transformers
Cenk Baykal, Dylan Cutler, Nishanth Dikkala +3
It has been well established that increasing scale in deep transformer networks leads to improved quality and performance. However, this increase in scale often comes with prohibit…