Xudong Shen

Ph.D. student

National University of Singapore

About Me : )

Hello! I am Xudong Shen (沈旭东). Friends also call me Oliver. I am a 5th year Ph.D. student at National Univeristy of Singapore, advised by professor Mohan Kankanhalli. I work on both the technical methods to ensure the social alignment of AI and the policy and regulation aspects of AI systems. With a particular focus on fairness, I have worked on measuring gender discrimination in P2P lending, general fair representation learning, and measuring & mitigating biases in large language models and text-to-image diffusion models. I can be reached at xudong.shen@u.nus.edu.

From 2022 to 2023, I interned at Sea AI Lab in Singapore. Between 2015 and 2019, I was an undergraduate at Zhejiang University, China, majoring in Naval Architect and Ocean Engineering. During this period, I explored various topics including computational fluid dynamics, satellite imaging, and policy research. My cv is here.

I believe in Diversity, Equity & Inclusion. I am committed to uphold these values.

If you are interested, I’ve written something more about myself…

Publications

[2024b] Xudong Shen, Hannah Brown, Jiashu Tao, Martin Strobel, Yao Tong, Akshay Narayan, Harold Soh, Finale Doshi-Velez, “Directions of Technical Innovation for Regulatable AI Systems”, Communications of the ACM, (forthcoming). [arXiv]

[2024a] Xudong Shen, Chao Du, Tianyu Pang, Min Lin, Yongkang Wong, and Mohan Kankanhalli, “Finetuning Text-to-Image Diffusion Models for Fairness”, ICLR, 2024. (Oral, top 1.2%). [arXiv] [Webpage]

[2023a] Xudong Shen, Tianhui Tan, Tuan Q. Phan, Jussi Keppo, “Gender Animus Can Still Exist Under Favorable Disparate Impact: a Cautionary Tale from Online P2P Lending”, ACM Conference on Fairness, Accountability, and Transparency (FAccT), 2023. [Paper] [arXiv] [Video]

[2023b] Xudong Shen, Yongkang Wong, and Mohan S Kankanhalli, “Fair Representation: Guaranteeing Approximate Multiple Group Fairness for Unknown Tasks”, IEEE Trans. PAMI, 2023. [Paper] [arXiv] [GitHub]

[2023c] Ian McKenzie, …, Xudong Shen, …, “Inverse Scaling: When Bigger Isn’t Better”, TMLR, 2023. [arXiv] [GitHub]

[2023d] Aarohi Srivastava, …, Xudong Shen, …, “Beyond the Imitation Game: Quantifying and Extrapolating the Capabilities of Language Models”, JMLR, 2023. [Paper] [arXiv]

[2023e] Kaustubh Dhole, …, Xudong Shen, …, “NL-Augmenter: A Framework for Task-Sensitive Natural Language Augmentation”, Northern European Journal of Language Technology (NEJLT), 2023. [arXiv] [GitHub]

[2022a] Yizhong Wang, …, Xudong Shen, …, “Benchmarking Generalization via In-Context Instructions on 1,600+ Language Tasks”, EMNLP, 2022. [Paper] [arXiv] [Data]

[2021a] Ziwei Xu, Xudong Shen, Yongkang Wong, and Mohan Kankanhalli, “Unsupervised Motion Representation Learning with Capsule Autoencoders”, NeurIPS, 2021. [Paper] [arXiv] [GitHub] [Slide] [Video]

Working Papers

[w1] Xudong Shen, Tianhui Tan, Jussi Keppo, and Tuan Q. Phan, Improved Identification of Gender Discrimination Drivers in Online P2P Lending.
— We show an improved identification of the taste-based and the belief-based gender discrimination drivers in P2P lending.
— Oral presentation at the International Conference on Smart Finance, 2022, [slides].
— Preparing journal submission.

[w2] Xudong Shen, Yongkang Wong, and Mohan Kankanhalli, Theoretical Guarantees of Subgroup-Fair Representation via Task Anticipation.

[w3] Ayse Gizem Yasar, Andrew Chong^†, Evan Dong^†, Thomas Krendl Gilbert^†, Sarah Hladikova^†, Roland Maio^†, Carlos Mougan^†, Xudong Shen^†, Shubham Singh^†, Ana-Andreea Stoica^†, Savannah Thais^†, ``Integration of Generative AI in the Digital Markets Act: Contestability and Fairness from a Cross-Disciplinary Perspective’’, (2024), under review for LSE working papers series. (^† = equal contribution.)