Shuoyang Ding bio photo

Shuoyang Ding

Email Twitter Github

About

I'm a researcher at NVIDIA that develops algorithms that process and generate information in both speech and text modalities.

I used to work as an applied scientist in Amazon AWS AI Labs, where I spent two years working on products such as Amazon Translate, Amazon Q for Business, and Amazon Titan Text Embeddings. Prior to Amazon, I was a PhD student at Department of Computer Science, Johns Hopkins University. My PhD advisor was Philipp Koehn.

During my graduate studies, I was primarily affiliated with Center for Language and Speech Processing and also did research on neural machine translation. I had also spent a few memorable months either interning at Microsoft Translator, Salesforce Research, Amazon, or visiting The University of Edinburgh. Before joining Johns Hopkins, I got my Bachelor's degree in Beijing University of Posts & Telecommunications. During the last year of my undergraduate study, I worked with Weiwei Sun in the Language Computing and Web Mining Group of Institute of Computer Science & Technology, Peking University, with a focus on semantic parsing and Chinese word segmentation.

Publications

  • Fine-Tuned Machine Translation Metrics Struggle in Unseen Domains
    Vilém Zouhar, Shuoyang Ding, Anna Currey, Tatyana Badeka, Jenyuan Wang, Brian Thompson
    Annual Meeting of the Association for Computational Linguistics (ACL) 2024 (To Appear) [pdf]

  • Doubly-Trained Adversarial Data Augmentation for Neural Machine Translation
    Weiting Tan, Shuoyang Ding, Huda Khayrallah, Philipp Koehn
    The 15th Conference of the Association for Machine Translation in the Americas (AMTA 2022) [pdf]

  • Runtime Audit of Neural Sequence Models for NLP
    Shuoyang Ding
    PhD Thesis [pdf]

  • The JHU-Microsoft Submission for WMT21 Quality Estimation Shared Task
    Shuoyang Ding, Marcin Junczys-Dowmunt, Matt Post, Christian Federmann, Philipp Koehn
    Sixth Conference on Machine Translation (WMT) 2021 [pdf][poster]

  • Levenshtein Training for Word-level Quality Estimation
    Shuoyang Ding, Marcin Junczys-Dowmunt, Matt Post, Philipp Koehn
    EMNLP 2021 [pdf][code][slides][poster][talk]

  • Evaluating Saliency Methods for Neural Language Models
    Shuoyang Ding, Philipp Koehn
    NAACL 2021 [pdf][code][slides][talk]

  • Espresso: A Fast End-to-end Neural Speech Recognition Toolkit
    Yiming Wang, Tongfei Chen, Hainan Xu, Shuoyang Ding, Hang Lv, Yiwen Shao, Nanyun Peng, Lei Xie, Shinji Watanabe, Sanjeev Khudanpur
    ASRU 2019 [pdf][code]

  • A Call for Prudent Choice of Subword Merge Operations in Neural Machine Translation
    Shuoyang Ding, Adithya Renduchintala, Kevin Duh
    MT Summit 2019 [pdf][code][poster]

  • An Exploration of Masking for Neural Machine Translation
    Matt Post, Shuoyang Ding, Marianna Martindale and Winston Wu
    MT Summit 2019 [pdf]

  • Saliency-driven Word Alignment Interpretation for Neural Machine Translation
    Shuoyang Ding, Hainan Xu, Philipp Koehn
    Fourth Conference on Machine Translation (WMT) 2019 [pdf][code][slides]

  • Parallelizable Stack Long Short-Term Memory
    Shuoyang Ding, Philipp Koehn
    NAACL 2019 Workshop on Structured Prediction for NLP [pdf][bib][code][slides]

  • Improving End-to-end Speech Recognition with Pronunciation-assisted Sub-word Modeling
    Hainan Xu, Shuoyang Ding, Shinji Watanabe
    ICASSP 2019 [pdf][bib][code]

  • Multi-Modal Data Augmentation for End-to-end ASR
    Adithya Renduchintala, Shuoyang Ding, Matthew Wiesner, Shinji Watanabe
    Interspeech 2018 Best Student Paper Award (3/700+) [pdf][bib]

  • The JHU Machine Translation Systems for WMT 2017
    Shuoyang Ding, Huda Khayrallah, Philipp Koehn, Matt Post, Gaurav Kumar, and Kevin Duh
    Second Conference on Machine Translation (WMT) 2017 [pdf][bib]

  • The JHU Machine Translation Systems for WMT 2016
    Shuoyang Ding, Kevin Duh, Huda Khayrallah, Philipp Koehn, and Matt Post
    First Conference on Machine Translation (WMT) 2016 [pdf][bib]

  • Grammatical Relations in Chinese: GB-Ground Extraction and Data-Driven Parsing
    Weiwei Sun, Yantao Du, Xin Kou, Shuoyang Ding, Xiaojun Wan
    Annual Meeting of the Association for Computational Linguistics (ACL) 2014 [pdf][bib]

Preprints

  • How Do Source-side Monolingual Word Embeddings Impact Neural Machine Translation?
    Shuoyang Ding and Kevin Duh, 2018 [pdf]

  • Backstitch: Counteracting Finite-sample Bias via Negative Steps
    Yiming Wang, Hossein Hadian, Shuoyang Ding, Ke Li, Hainan Xu, Xiaohui Zhang, Daniel Povey, Sanjeev Khudanpur, 2017 [pdf]

Courses

Teaching