Yi Su (苏毅)

prof_pic.jpg

1 Apple Park Way

Cupertino, CA 95014

I am a seasoned research engineer who possesses a unique combination of skill sets: machine learning research and software engineering.

With a solid education of PhD in speech and language processing and 10+ years of experience on research and development in an industrial setting, I have shipped numerous machine learning features for Apple Siri, Alipay Intelligent Assistant (“支付宝智能助理”), Ant Fortune Intelligent Financial Assistant (“支小宝”), Nuance Dragon NaturallySpeaking, Nuance PowerScribe 360 and Nuance Nina Virtual Assistant.

I have also built many internal software and services that are still in use today, including a large-scale n-gram language model trainer, a low-code chatbot development platform and an offline model evaluation framework.

In general, I am interested in statistical modeling of human intelligence as exemplified in the use of language and speech.

I like traveling.

news

Dec 30, 2024 Check out our ACL Workshop paper and NeurIPS Workshop paper!
Mar 13, 2024 Rebuilt my personal website with al-folio and it looks great!
Feb 16, 2024 Check out our ICASSP ‘24 paper!
Jun 09, 2023 Check out my pet project LLaMA Server!
Oct 01, 2022 Check out our JMLR paper and code!

selected publications

  1. NeurIPS
    Device-Directed Speech Detection for Follow-up Conversations Using Large Language Models
    Ognjen Rudovic , Pranay Dighe , Yi Su, and 6 more authors
    In The Thirty-Eighth Annual Conference on Neural Information Processing Systems Workshop on Adaptive Foundation Models , 2024
    PDF
  2. ACL
    KGLens: Towards Efficient and Effective Knowledge Probing of Large Language Models with Knowledge Graphs
    Shangshang Zheng , He Bai , Yizhe Zhang , and 3 more authors
    In The 62nd Annual Meeting of the Association for Computational Linguistics Workshop on Towards Knowledgeable Language Models , 2024
    PDF
  3. ICASSP
    Leveraging large language models for exploiting ASR uncertainty
    Pranay Dighe , Yi Su, Shangshang Zheng , and 4 more authors
    In IEEE International Conference on Acoustics, Speech and Signal Processing , 2024
    PDF Poster
  4. JMLR
    Tianshou: A highly modularized deep reinforcement learning library
    Jiayi Weng , Huayu Chen , Dong Yan , and 6 more authors
    Journal of Machine Learning Research, 2022
    PDF Code