Learning with Imperfect Supervision
We design algorithms for noisy labels, coarse labels, open-set noise, and other forms of imperfect supervision in real-world visual datasets.
Our work is motivated by a simple problem: high-performing AI systems are often trained and evaluated with imperfect information. We develop methods that make learning, evaluation, and reasoning more reliable in those conditions.
We design algorithms for noisy labels, coarse labels, open-set noise, and other forms of imperfect supervision in real-world visual datasets.
We study certification, adversarial robustness, and statistically reliable evaluation for models and LLM-based systems under imperfect judges.
We adapt foundation models such as CLIP-style representations to downstream settings where data quality, annotation quality, or domain fit is limited.
We analyze failure modes of idealized noise correction and build practical robust learning methods for noisy supervision.
We work on robust statistical evaluation protocols for LLMs and foundation models when automatic or human judges are noisy.
We develop certification methods that provide formal or statistical guarantees under adversarial perturbations and distribution shifts.