Kumari Kavita

Alessandro Pegoraro, Kavita Kumari, Hossein Fereidooni et al.

ChatGPT has become a global sensation. As ChatGPT and other Large Language Models (LLMs) emerge, concerns of misusing them in various ways increase, such as disseminating fake news, plagiarism, manipulating public opinion, cheating, and fraud. Hence, distinguishing AI-generated from human-generated becomes increasingly essential. Researchers have proposed various detection methodologies, ranging from basic binary classifiers to more complex deep-learning models. Some detection techniques rely on statistical characteristics or syntactic patterns, while others incorporate semantic or contextual information to improve accuracy. The primary objective of this study is to provide a comprehensive and contemporary assessment of the most recent techniques in ChatGPT detection. Additionally, we evaluated other AI-generated text detection tools that do not specifically claim to detect ChatGPT-generated content to assess their performance in detecting ChatGPT-generated content. For our evaluation, we have curated a benchmark dataset consisting of prompts from ChatGPT and humans, including diverse questions from medical, open Q&A, and finance domains and user-generated responses from popular social networking platforms. The dataset serves as a reference to assess the performance of various techniques in detecting ChatGPT-generated content. Our evaluation results demonstrate that none of the existing methods can effectively detect ChatGPT-generated content.

Kavita Kumari, Murtuza Jadliwala, Sumit Kumar Jha et al.

Model explanations improve the transparency of black-box machine learning (ML) models and their decisions; however, they can also be exploited to carry out privacy threats such as membership inference attacks (MIA). Existing works have only analyzed MIA in a single "what if" interaction scenario between an adversary and the target ML model; thus, it does not discern the factors impacting the capabilities of an adversary in launching MIA in repeated interaction settings. Additionally, these works rely on assumptions about the adversary's knowledge of the target model's structure and, thus, do not guarantee the optimality of the predefined threshold required to distinguish the members from non-members. In this paper, we delve into the domain of explanation-based threshold attacks, where the adversary endeavors to carry out MIA attacks by leveraging the variance of explanations through iterative interactions with the system comprising of the target ML model and its corresponding explanation method. We model such interactions by employing a continuous-time stochastic signaling game framework. In our framework, an adversary plays a stopping game, interacting with the system (having imperfect information about the type of an adversary, i.e., honest or malicious) to obtain explanation variance information and computing an optimal threshold to determine the membership of a datapoint accurately. First, we propose a sound mathematical formulation to prove that such an optimal threshold exists, which can be used to launch MIA. Then, we characterize the conditions under which a unique Markov perfect equilibrium (or steady state) exists in this dynamic system. By means of a comprehensive set of simulations of the proposed game model, we assess different factors that can impact the capability of an adversary to launch MIA in such repeated interaction settings.

Alessandro Pegoraro, Carlotta Segna, Kavita Kumari et al.

Deep Learning (DL) models have become crucial in digital transformation, thus raising concerns about their intellectual property rights. Different watermarking techniques have been developed to protect Deep Neural Networks (DNNs) from IP infringement, creating a competitive field for DNN watermarking and removal methods. The predominant watermarking schemes use white-box techniques, which involve modifying weights by adding a unique signature to specific DNN layers. On the other hand, existing attacks on white-box watermarking usually require knowledge of the specific deployed watermarking scheme or access to the underlying data for further training and fine-tuning. We propose DeepEclipse, a novel and unified framework designed to remove white-box watermarks. We present obfuscation techniques that significantly differ from the existing white-box watermarking removal schemes. DeepEclipse can evade watermark detection without prior knowledge of the underlying watermarking scheme, additional data, or training and fine-tuning. Our evaluation reveals that DeepEclipse excels in breaking multiple white-box watermarking schemes, reducing watermark detection to random guessing while maintaining a similar model accuracy as the original one. Our framework showcases a promising solution to address the ongoing DNN watermark protection and removal challenges.