1

we develop a novel framework for crafting generalizable backdoors in RLHF through emotion-aware trigger synthesis

We introduce Generative Adversarial Suffix Prompter (GASP), a novel framework that combines human-readable prompt generation with Latent Bayesian Optimization (LBO) to improve adversarial suffix creation in a fully black-box setting.

We propose a multi-agent, multi-turn jailbreak strategy that systematically bypasses LLM safety mechanisms by decomposing harmful queries into seemingly benign sub-tasks.

We develop a novel attack framework that generates highly invisible adversarial patches based on perceptibility-aware localization and perturbation schemes.

We study how to certify and train for cost-sensitive robustness using randomized smoothing.

We propose DiffPAD, a novel framework that harnesses the power of diffusion models for adversarial patch decontamination.

We highlight the importance of using dynamic FR strategies to evaluate AFR methods, and propose DivTrackee as a promising countermeasure.

Understand the inherent vulnerabilities to indiscriminate data poisoning attacks for linear learners by studying the optimal poisoning strategy from the perspective of data distribution.

Built upon a novel definition of label uncertainty, we develop an empirical method to estimate a more realistic intrinsic robustness limit for image classification tasks.

We show that concentration of measure does not prohibit the existence of adversarially robust classifiers using a novel method of empirical concentration estimation.