Photo of Charlie Lu

Charlie Lu

PhD student, MIT Media Lab

About

I'm a PhD student in the Camera Culture group at the MIT Media Lab, advised by Ramesh Raskar and supported by an NSF Graduate Research Fellowship. I study trustworthy knowledge agents: when we can trust what LLM agents report. Previously, I worked on conformal prediction and data markets, and built medical imaging models at Massachusetts General Hospital, including an FDA-cleared stroke detection model.

I'm looking for research internships for summer 2027.

Outside the lab, I enjoy woodworking, < href="https://arts.mit.edu/inside-the-saa-ceramics-studio/">pottery, and digital fabrication.

Selected Publications

Right Number, Wrong Question: LLM Probability Substitution in Incomplete Searches
Charles Lu, A. Molina, R. Raskar. Under review at ICLR 2027.
Asking for Trouble: The Help Channel as an Injection Surface
Charles Lu, J. Dunfoy, A. Borra, D. Shende, G. Grant-Richards, J. Li, J. She, T. Klinghoffer, C. Wang, U. Madhushani Sehwag. Under review at ICLR 2027.
Data Acquisition via Experimental Design for Data Markets
Charles Lu, B. Huang, S. P. Karimireddy, P. Vepakomma, M. I. Jordan, R. Raskar. NeurIPS 2024. [arXiv]
Federated Conformal Predictors for Distributed Uncertainty Quantification
Charles Lu*, Y. Yu*, S. P. Karimireddy, M. I. Jordan, R. Raskar. ICML 2023.
Conformal Prediction with Large Language Models for Multi-Choice Question Answering
B. Kumar*, Charles Lu*, G. Gupta, A. Palepu, D. Bellamy, R. Raskar, A. Beam. ICML 2023 Workshop.
Improving Trustworthiness of AI Disease Severity Rating in Medical Imaging with Ordinal Conformal Prediction Sets
Charles Lu*, A. N. Angelopoulos*, S. Pomerantz. MICCAI 2022.
Fair Conformal Predictors for Applications in Medical Imaging
Charles Lu, A. Lemay, K. Chang, K. Hoebel, J. Kalpathy-Cramer. AAAI 2022.

* Equal contribution. Full list on Google Scholar.

Experience