Deric Pang

I work on AI for design at Figma in NYC.

Previously, I worked on semantic parsing and question answering for Google Search, reinforcement learning in simulation at Unity Technologies, training autonomous robots in simulation at NVIDIA, and a deep learning framework for speech recognition at Amazon Alexa AI.

I got my B.S. and M.S. in computer science at the University of Washington, where I did research in Natural Language Processing with Noah Smith.

Blog

Sim2Real
Training a rover to navigate an indoor office environment with a simulator.
- Video
- Slides
Tellina
Translating plain English into Bash commands using deep learning.
- Paper
- Code
Syntactic Entailment
Improving natural language inference with syntactic word representations.
- Paper
- Code

Improving Natural Language Inference with a Pretrained Parser
Deric Pang, Lucy H. Lin, and Noah A. Smith
arXiv:1909.08217, 2019
Improving Natural Language Inference with Syntactic Word Embeddings
Deric Pang
Master's Thesis, University of Washington, 2019
- PDF
- Code
Program Synthesis from Natural Language Using Recurrent Neural Networks
Xi Victoria Lin, Chenglong Wang, Deric Pang, Kevin Lu, Luke Zettlemoyer, and Michael D. Ernst
Technical Report, University of Washington, 2017
- PDF
- Code
Evaluating and improving fault localization
Spencer Pearson, José Campos, René Just, Gordon Fraser, Rui Abreu, Michael D. Ernst, Deric Pang, and Benjamin Keller
2017 IEEE/ACM 39th International Conference on Software Engineering (ICSE)
- ICSE
- PDF