Publications by Matt Fredrikson

Conference

AGENTHARM: A BENCHMARK FOR MEASURING HARMFULNESS OF LLM AGENTS

2025 • 13th International Conference on Learning Representations Iclr 2025 • 18136-18171
Andriushchenko M, Souly A, Dziemian M, Duenas D, Lin M, Wang J, Hendrycks D, Zou A, Kolter Z, Fredrikson M, Gal Y, Davies X

Conference

ALIGNED LLMS ARE NOT ALIGNED BROWSER AGENTS

2025 • 13th International Conference on Learning Representations Iclr 2025 • 62386-62407
Kumar P, Lau E, Vijayakumar S, Trinh T, Team SR, Chang E, Robinson V, Hendryx S, Zhou S, Fredrikson M, Yue S, Wang Z

Conference

A RECIPE FOR IMPROVED CERTIFIABLE ROBUSTNESS

2024 • 12th International Conference on Learning Representations Iclr 2024
Hu K, Leino K, Wang Z, Fredrikson M

Conference

Attacks and Defenses for Large Language Models on Coding Tasks

2024 • Proceedings / IEEE International Conference, Automated Software Engineering ; sponsored by IEEE Computer Society, NASA Ames Research Center, in cooperation with AAAI, ACM SIGART and SIGSOFT. IEEE International Automated Software Enginee... • 2268-2272
Zhang C, Wang Z, Zhao R, Mangal R, Fredrikson M, Jia L, Pasareanu CS

Conference

ON THE PERILS OF CASCADING ROBUST CLASSIFIERS

2023 • 11th International Conference on Learning Representations Iclr 2023
Mangal R, Wang Z, Zhang C, Leino K, Păsăreanu C, Fredrikson M

Conference

CONSISTENT COUNTERFACTUALS FOR DEEP MODELS

2022 • Iclr 2022 10th International Conference on Learning Representations
Black E, Wang Z, Datta A, Fredrikson M

Journal Article

Degradation Attacks on Certifiably Robust Neural Networks

2022 • Transactions on Machine Learning Research • 1(1):
Leino K, Zhang C, Mangal R, Fredrikson M, Parno B, Pasareanu C