Name: Hua Shen
Type: User
Company: University of Michigan
Bio: Postdoc @UMich. My research interests include human-centered AI interpretability and fairness, speech and NLP, human-AI interaction.
Twitter: huashen218
Location: Ann Arbor, MI
Blog: http://hua-shen.org
Hua Shen's Projects
Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them
Beyond Accuracy: Behavioral Testing of NLP models with CheckList
CSCW 2023 Demo paper: Conversational AI Explanations to Support Human-AI Scientific Writing
HTML pattern library for CTZN
A benchmark for understanding and evaluating rationales: http://www.eraserbenchmark.com/
Implementation for https://arxiv.org/abs/2005.00652
HCOMP 2020 paper - human eval on XAI (saliency maps for computer vision models)
Lime: Explaining the predictions of any machine learning classifier
ACL 2022 paper - human eval on self-explaining models
HCOMP 2023 paper - human collaboration on detecting LLM-generated deepfake texts
Code samples to help you get started with the Amazon Mechanical Turk Requester API
Companion site for "Analysis Methods in Neural Language Processing: A Survey"
💬 Open source machine learning framework to automate text- and voice-based conversations: NLU, dialogue management, connect to Slack, Facebook, and more - Create chatbots and voice assistants
Robustness Gym: Evaluation Toolkit for NLP
A game theoretic approach to explain the output of any machine learning model.
An Exploration of Prompt Tuning on Generative Spoken Language Model for Speech Processing Tasks (INTERSPEECH 2022)
TalkToModel gives anyone with the powers of XAI through natural language conversations 💬!
TrojanZoo provides a universal pytorch platform to conduct security researches (especially backdoor attacks/defenses) of image classification in deep learning.
ICASSP 2022 paper - improve fairness of speaker verification models