Aaron Chan

Research Scientist at Meta AI   •   aarzchan :slightly_smiling_face: gmail :upside_down_face: com


prof_pic.jpg

I am a research scientist on the AI for Modern Recommendation Systems (MRS) Team at Meta AI.

My research is at the intersection of natural language processing and machine learning. In particular, I am excited about:

  • model explainability: explaining language model behavior more faithfully, plausibly, and efficiently.
  • explanation-based learning: operationalizing explanations to improve language model generalization and decision-making.

Previously, I was a research intern at Meta AI and an engineering intern at Google.

I earned my PhD in computer science from the University of Southern California. During my PhD, I was advised by Prof. Xiang Ren at the INK Lab. Before that, I earned my MSE in robotics from the University of Pennsylvania and my BS in electrical engineering from the University of Maryland.

Currently, I work remotely from the Washington, DC metro area, where I live with my wife. Outside of work, I enjoy basketball, skiing, hiking, reading, and board games.


News

Mar 10, 2023 HUFTR was accepted to the TRAIT Workshop at CHI 2023! :tada:
Mar 04, 2023 KNIFE was accepted to the TrustML-(un)Limited Workshop at ICLR 2023! :tada:
Feb 28, 2023 Check out our latest version of ER-Test on arXiv, updated with additional experiments and improved presentation! :page_facing_up:
Feb 27, 2023 Check out our latest version of UNIREX on arXiv, updated with additional experiments and improved presentation! :page_facing_up:
Feb 14, 2023 Our code for UNIREX has been officially released on the Meta Research GitHub! :computer:
See all news >

Selected Publications

  1. ICLR
    PINTO: Faithful Language Reasoning Using Prompt-Generated Rationales
    P. Wang, A. Chan, F. Ilievski, M. Chen, and X. Ren
    ICLR, 2023
    arXiv PDF Code
  2. EMNLP
    ER-Test: Evaluating Explanation Regularization Methods for Language Models
    B. Joshi*, A. Chan*, Z. Liu*, S. Nie, M. Sanjabi, H. Firooz, and X. Ren
    Findings of EMNLP, 2022
    arXiv PDF Code
  3. ICML
    UNIREX: A Unified Learning Framework for Language Model Rationale Extraction
    A. Chan, M. Sanjabi, L. Mathias, L. Tan, S. Nie, X. Peng, X. Ren, and H. Firooz
    ICML, 2022
    arXiv PDF Code
  4. NeurIPS
    SalKG: Learning From Knowledge Graph Explanations for Commonsense Reasoning
    A. Chan, J. Xu, B. Long, S. Sanyal, T. Gupta, and X. Ren
    NeurIPS, 2021
    arXiv PDF Code
  5. ICLR
    Learning to Deceive Knowledge Graph Augmented Models via Targeted Perturbation
    M. Raman, A. Chan*, S. Agarwal*, P. Wang, H. Wang, S. Kim, R. Rossi, H. Zhao, N. Lipka, and X. Ren
    ICLR, 2021
    arXiv PDF Code