EdgeBERT: Optimizing On-Chip Inference for Multi-Task NLP

Thierry Tambe
Thierry Tambe
Coleman Hooper
Coleman Hooper
Lillian Pentecost
Lillian Pentecost
En-Yu Yang
En-Yu Yang
Victor Sanh
Victor Sanh
Cited by: 0|Bibtex|Views28
Other Links: arxiv.org

Abstract:

Transformer-based language models such as BERT provide significant accuracy improvement to a multitude of natural language processing (NLP) tasks. However, their hefty computational and memory demands make them challenging to deploy to resource-constrained edge platforms with strict latency requirements. We present EdgeBERT an in-dept...More

Code:

Data:

Full Text
Your rating :
0

 

Tags
Comments