Ashia C. Wilson
Lister Brothers Career Development Associate Professor, MIT
MIT EECS, LIDS, AI for Society, SDSC
ashia07@mit.edu · Stata Center · Google Scholar · GitHub
Research
Dr. Ashia Wilson runs the Optimization, Safety, and Evaluation (OSE) Lab. The lab develops methods and foundations for AI safety, focusing on image-based abuse and child safety, and on understanding and mitigating epistemic harms such as homogenization and sycophancy.
Recent News
- I am co-organizing a workshop at NeurIPS this year on child safety.
- “Evaluation without Generation” received a Spotlight at the AI4GOOD Workshop and was highlighted in MIT News.
- I co-organized a tutorial at ICML on unlearning.
- Our work studying sycophancy received an honorable mention at this year's CHI Conference and was highlighted in MIT News.
- I was named Junior Bose Award winner.
- Multiple recent papers accepted to ICML, AISTATS, CHI, FAccT, COLT, and TMLR.
- “Near-Optimal Private Linear Regression via Iterative Hessian Mixing” received an ICML Spotlight.