Ashia C. Wilson

Lister Brothers Career Development Assistant Professor, MIT

For a full list of publications, see Google Scholar.

Ashia C. Wilson

Bio

I obtained my B.A. from Harvard with a concentration in applied mathematics and a minor in philosophy, and my Ph.D. from UC Berkeley in statistics. Before joining MIT, I held a postdoctoral position in the machine learning and statistics group at Microsoft Research.

Affiliations

  • MIT Department of Electrical Engineering and Computer Science (EECS)
  • MIT Lab for Information and Decision Systems (LIDS)
  • MIT Statistics and Data Science Center (SDSC)

Recent Publications & Working Papers

Evaluation

Aligning Evaluation with Clinical Priorities: Calibration, Label Shift, and Error Costs

Joint with Gerardo Flores, Alyssa Smith, and Julia Fukuyama

NeurIPS 2025 [arXiv]
Unlearning & Privacy

The Gaussian Mixing Mechanism: Renyi Differential Privacy via Gaussian Sketches

Joint with Omri Lev, Vishwak Srinivasan, Moshe Shenfeld, Katrina Ligett, and Ayush Sekhari

NeurIPS 2025 [arXiv]
Homogenization

Homogeneous Algorithms Can Reduce Competition in Personalized Pricing

Joint with Nathan Jo, Kathleen Creel, and Manish Raghavan

NeurIPS 2025 [arXiv]
Optimization & Sampling

Adaptive Backtracking Line Search

Joint with Laurent Lessard and Joao Cavalcanti

ICLR 2025 [arXiv]
Optimization & Sampling

High-accuracy sampling from constrained spaces with the Metropolis-adjusted Preconditioned Langevin Algorithm

Joint with Vishwak Srinivasan and Andre Wibisono

ALT 2025 [arXiv]
Homogenization

Allocation Multiplicity: Evaluating the Promises of the Rashomon Set

Joint with Shomik Jain, Margaret Wang, and Kathleen Creel

FAccT 2025 [arXiv]
Homogenization

Algorithmic Pluralism: A Structural Approach To Equal Opportunity 🏆 Best Paper

Joint with Shomik Jain, Vinith Suriyakumar, and Kathleen Creel

FAccT 2024 [arXiv]
Homogenization

Scarce Resource Allocations That Rely On Machine Learning Should Be Randomized

Joint with Shomik Jain and Kathleen Creel

ICML 2024 [arXiv]
Optimization & Sampling

Adaptive kernel selection for Stein Variational Gradient Descent

Joint with Moritz Melcher, Simon Weissmann, and Jakob Zech

Preprint 2025 [arXiv]
Evaluation

Interaction Context Often Increases Sycophancy in LLMs

Joint with Shomik Jain, Charlotte Park, Matheus Vianna, and Dana Calacci

Preprint 2025 [arXiv]
Evaluation

A Consequentialist Critique of Binary Classification Evaluation: Theory, Practice, and Tools

Joint with G Flores, A Schiff, AH Smith, JA Fukuyama

Preprint 2025 [arXiv]
Unlearning & Privacy

Unstable Unlearning: The Hidden Risk of Concept Resurgence in Diffusion Models

Joint with Vinith Suriyakumar, Rohan Alur, Ayush Sekhari and Manish Raghavan

Preprint 2024 [arXiv]
Unlearning & Privacy

UCD: Unlearning in LLMs via Contrastive Decoding

Joint with Vinith M. Suriyakumar and Ayush Sekhari

Preprint 2025 [arXiv]
Evaluation

What Does Your Benchmark Really Measure? A Framework for Robust Inference of AI Capabilities

Joint with Nathan Jo

Preprint 2024 [arXiv]
Homogenization

LLM Output Homogenization is Task Dependent

Joint with Shomik Jain, Jack Lanchantin, Maximilian Nickel, Karen Ullrich and Jamelle Watson-Daniels

Preprint 2024 [arXiv]

Teaching

  • 6.401
    Introduction to Data Science and Statistics

    Spring 2021

  • 6.036
    Introduction to Machine Learning

    Spring 2022, Spring 2023

  • 6.3950
    AI, Decision Making and Society

    Fall 2022, Fall 2023, Fall 2024, Fall 2025

  • 6.S964
    Topics in Data Science for Society

    Spring 2025

#

Contact

Contact Information

Academic Links