| CARVIEW |
aarzchan 🙂 gmail 🙃 com
Bio
I am a research scientist at Meta AI. Broadly, my research interests are at the intersection of AI, ML, and NLP. In particular, I am excited about generative AI and LLMs, especially in the context of AI safety and AI alignment.
Currently, I am on the Meta AI Safety Team within Meta's GenAI org. As a member of this team, I work on developing LLM post-training algorithms to make the Meta AI assistant more helpful, honest, and harmless. Our work directly impacts over 600M global users across Meta.AI, Facebook, Instagram, Messenger, and WhatsApp.
Prior to GenAI, I was part of the Modern Recommendation Systems (MRS) org at Meta. In MRS, I worked on building ML models to improve Reels ranking on Facebook and Instagram, with a focus on long-term user value optimization. Previously, I was a research intern on the AI Integrity Team at Meta, where I worked on LLM explainability. Before that, I was an engineering intern on the Android Camera Team at Google, where I worked on image saliency detection.
I earned my PhD in computer science from the University of Southern California, advised by Prof. Xiang Ren at INK Lab. During my PhD, I conducted fundamental research in ML and NLP, developing algorithms for model explainability and explanation-based learning. I earned my MSE in robotics from the University of Pennsylvania and my BS in electrical engineering from the University of Maryland.
Outside of work, I enjoy tennis, basketball, pickleball, skiing, and spending time with my family.Publications
* = equal contribution- ResPrompt: Residual Connection Prompting Advances Multi-Step Reasoning in Large Language Models
S. Jiang, Z. Shakeri, A. Chan, M. Sanjabi, H. Firooz, Y. Xia, B. Akyildiz, Y. Sun, J. Li, Q. Wang, A. Celikyilmaz
NAACL 2024
[Paper] - Tailoring Self-Rationalizers with Multi-Reward Distillation
S. Ramnath, B. Joshi, S. Hallinan, X. Lu, L. Li, A. Chan, J. Hessel, Y. Choi, X. Ren
ICLR 2024 | SeT LLM Workshop at ICLR 2024
[Paper] [Code] [Website] - KNIFE: Distilling Reasoning Knowledge From Free-Text Rationales
A. Chan*, Z. Zeng*, W. Lake, B. Joshi, H. Chen, X. Ren
TrustML-(un)Limited Workshop at ICLR 2023
[Paper] - XMD: An End-to-End Framework for Interactive Explanation-Based Debugging of NLP Models
D. Lee*, A. Kadakia*, B. Joshi, A. Chan, Z. Liu, K. Narahari, T. Shibuya, R. Mitani, T. Sekiya, J. Pujara, X. Ren
ACL 2023 - Demo Track
[Paper] [Code] [Website] - Are Machine Rationales (Not) Useful to Humans? Measuring and Improving Human Utility of Free-Text Rationales
B. Joshi*, Z. Liu*, S. Ramnath, A. Chan, Z. Tong, Q. Wang, Y. Choi, X. Ren
ACL 2023 (Oral) | TRAIT Workshop at CHI 2023
[Paper] [Code] - PINTO: Faithful Language Reasoning Using Prompt-Generated Rationales
P. Wang, A. Chan, F. Ilievski, M. Chen, X. Ren
ICLR 2023 | TL4NLP Workshop at NeurIPS 2022 | TSRML Workshop at NeurIPS 2022
[Paper] [Code] - FRAME: Evaluating Rationale-Label Consistency Metrics for Free-Text Rationales
A. Chan, S. Nie, L. Tan, X. Peng, H. Firooz, M. Sanjabi, X. Ren
BlackboxNLP Workshop at EMNLP 2022
[Paper] - ER-Test: Evaluating Explanation Regularization Methods for Language Models
B. Joshi*, A. Chan*, Z. Liu*, S. Nie, M. Sanjabi, H. Firooz, X. Ren
Findings of EMNLP 2022 | TrustNLP Workshop at NAACL 2022
[Paper] [Code] - UNIREX: A Unified Learning Framework for Language Model Rationale Extraction
A. Chan, M. Sanjabi, L. Mathias, L. Tan, S. Nie, X. Peng, X. Ren, H. Firooz
ICML 2022 (Spotlight) | SRML Workshop at ICLR 2022 | BigScience Workshop at ACL 2022
[Paper] [Code] - SalKG: Learning From Knowledge Graph Explanations for Commonsense Reasoning
A. Chan, J. Xu, B. Long, S. Sanyal, T. Gupta, X. Ren
NeurIPS 2021 | XAI Workshop at ICML 2021
[Paper] [Code] - Learning Contextualized Knowledge Structures for Commonsense Reasoning
J. Yan, M. Raman, A. Chan, T. Zhang, R. Rossi, H. Zhao, S. Kim, N. Lipka, X. Ren
Findings of ACL 2021 | KR2ML Workshop at NeurIPS 2020
[Paper] [Code] - 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, X. Ren
ICLR 2021 | KR2ML Workshop at NeurIPS 2020 (Best Paper Award Finalist)
[Paper] [Code] - Egocentric Basketball Motion Planning from a Single First-Person Image
G. Bertasius, A. Chan, J. Shi
CVPR 2018 | MIT SSAC 2018
[Paper] - 6-DoF Object Pose from Semantic Keypoints
G. Pavlakos, X. Zhou, A. Chan, K. Derpanis, K. Daniilidis
ICRA 2017
[Paper] [Code] - Scalable Vision System for Mouse Homecage Ethology
G. Salem, J. Krynitsky, B. Kirkland, E. Lin, A. Chan, S. Anfinrud, S. Anderson, M. Garmendia-Cedillos, R. Belayachi, J. Alonso-Cruz, J. Yu, A. Iano-Fletcher, G. Dold, T. Talbot, A.V. Kravitz, J.B. Mitchell, G. Wu, J.U. Dennis, M. Hayes, K. Branson, T. Pohida
ACIVS 2016
[Paper] [Website]