• Publications
  • Influence
MobiRNN: Efficient Recurrent Neural Network Execution on Mobile GPU
TLDR
We present MobiRNN, a mobile specific optimization for RNNs that focusses on offloading deep learning tasks to the mobile GPU. Expand
  • 33
  • 4
  • PDF
UIWear: Easily Adapting User Interfaces for Wearable Devices
TLDR
We present a programming model for wearable devices that abstracts a logical model of the smartphone GUI, re-tailors the GUI for the wearable device based on the specified UI design, and compiles it into a companion app that we call UICompanion app. Expand
  • 13
  • PDF
DeQA: On-Device Question Answering
TLDR
We present DeQA, a suite of latency- and memory- optimizations that adapts existing QA systems to run completely locally on mobile phones. Expand
  • 4
  • PDF
DeFormer: Decomposing Pre-trained Transformers for Faster Question Answering
TLDR
We introduce DeFormer, a decomposed transformer, which substitutes the full self-attention with question-wide and passage-wide selfattentions in the lower layers, which in turn enables pre-computing passage representations reducing runtime compute drastically. Expand
  • 3
  • PDF
DeQA: On-DeviceQuestion Answering
Today there is no effective support for device-wide question answering on mobile devices. State-of-the-art QAmodels are deep learning behemoths designed for the cloud which run extremely slow andExpand
Demo: UIWear: Easily Adapting User Interfaces for Wearable Devices
TLDR
We present a new programming model for wearable devices, where the developer writes the smartphone application once, and writes a simple meta program to encode the GUI design for the wearable device; no effort is needed beyond the design phase. Expand
Towards Accurate and Reliable Energy Measurement of NLP Models
TLDR
We quantify the error of existing software-based energy measurements by using a hardware power meter that provides highly accurate energy measurements. Expand