6 accepted papers
With the increased deployment of large language models (LLMs), one concern is their potential misuse for generating harmful content. Our work studies the alignment challenge, with a focus on filters to prevent the generation of unsafe information. Two natural points of intervention are the filtering…
We formalize and analyze the trade-off between backdoor-based watermarks and adversarial defenses, framing it as an interactive protocol between a verifier and a prover. While previous works have primarily focused on this trade-off, our analysis extends it by identifying transferable attacks as a th…
A new model for adversarial robustness was introduced by Goldwasser et al. in [GKKM20]. In this model the authors present a selective and transductive learning algorithm which guarantees a low test error and low rejection rate wrt to the original distribution. Moreover, a lower bound in terms of the…
We prove an exponential separation for the sample/query complexity between the standard PAC-learning model and a version of the Equivalence-Query-learning model. In the PAC model all samples are provided at the beginning of the learning process. In the Equivalence-Query model the samples are acquire…
There are two main attack models considered in the adversarial robustness literature: black-box and white-box. We consider these threat models as two ends of a fine-grained spectrum, indexed by the number of queries the adversary can ask. Using this point of view we investigate how many queries the…
Modern machine learning models with very high accuracy have been shown to be vulnerable to small, adversarially chosen perturbations of the input. Given black-box access to a high-accuracy classifier f, we show how to construct a new classifier g that has high accuracy and is also robust to adversar…