Paper “DRAGON: Distributional Rewards Optimize Diffusion Generative Models” to appear in Transactions on Machine Learning Research.
Three papers accepted at NeurIPS 2025: 1) Generalization or Hallucination? Understanding Out-of-Context Reasoning in Transformers, 2) OVERT: A Benchmark for Over-Refusal Evaluation on Text-to-Image Models, and 3) Revising and Falsifying Sparse Autoencoder Feature Explanations.
Four new PhD students have joined her research group: Alex Zalles, Kaiwen Hu, Srikar Babu Gadipudi, and Xutao Ma.
Paper “Evolutionary games on infinite strategy sets: Convergence to Nash equilibria via dissipativity” to appear in The IEEE Transactions on Automatic Control.
Paper “Coordinating Distributed Energy Resources with Nodal Pricing in Distribution Networks: a Game-Theoretic Approach” to appear in the 2025 IEEE Conference on Decision and Control.
Paper “Intent Demonstration in General-Sum Dynamic Games via Iterative Linear-Quadratic Approximations” has been conditionally accepted for publication in IEEE Transactions on Control Systems Technology (TCST).
Paper “OVERT: A Benchmark for Over-Refusal Evaluation on Text-to-Image Model” has been accepted for presentation at the MUGen Workshop, ICML 2025.
Paper “Towards optimal branching of linear and semidefinite relaxations for neural network robustness certification” to appear in the Journal of Machine Learning Research (JMLR).
Three PhD students from her group (Yatong Bai, Samuel Pfrommer, and Jingqi Li) have completed their dissertation talks.
PhD student, Jingqi Li (co-advised with Prof. Tomlin), received the 2024-2025 Leon O. Chua Award for outstanding achievement in nonlinear science from UC Berkeley.
PhD student, Yatong Bai, received the 2024-2025 Outstanding Graduate Student Instructor Award from UC Berkeley.
Paper “Distribution System Expansion Planning with Nonlinear Power Flow Models” to appear in IEEE PES General Meeting.
Appointed as the chair of the Women in Systems Engineering (WiSE) Committee of the IEEE Systems Council.
Appointed as an Associate Editor for Book Reviews of the IEEE Control Systems Magazine.
Paper “Certifiable Reachability Learning Using a New Lipschitz Continuous Value Function” has been accepted for publication in the IEEE Robotics and Automation Letters.
Paper “Distributionally Robust Joint Chance-Constrained Optimal Power Flow Using Relative Entropy” has been accepted for publication in the IEEE Transactions on Power Systems.
Paper “Ranking Manipulation for Conversational Search Engines” has been accepted for presentation at the EMNLP 2024 (main conference).
Research Experience
Serves as an Associate Professor at UC Berkeley, mentoring multiple PhD students in research work.
Education
No specific educational background information provided such as degrees, schools, advisors, etc.
Background
Associate Professor in the Electrical Engineering & Computer Sciences and Mechanical Engineering departments at UC Berkeley. Works on interdisciplinary problems in optimization theory, machine learning, power systems, and control.
Miscellany
Has a strong interest in optimization theory, machine learning, power systems, and control.