4 accepted papers
Foundation models are increasingly becoming better autonomous programmers, raising the prospect that they could also automate dangerous offensive cyber‑operations. Current frontier model audits probe the cybersecurity risks of such agents, but most fail to account for the degrees of freedom availabl…
Large Language Models (LLMs) have shown impressive capabilities in language understanding and generation, leading to growing interest in zero-shot relation triplet extraction (ZeroRTE), a task that aims to extract triplets for unseen relations without annotated data. However, existing methods typica
Although the prevention of AI vulnerabilities is critical to preserve the safety and privacy of users and businesses, educational tools for robust AI are still underdeveloped worldwide. We present the design, implementation, and assessment of Maestro. Maestro is an effective open-source game-based p…
Natural language processing (NLP) models are known vulnerable to adversarial examples, similar to image processing models. Studying adversarial texts is an essential step to improve the robustness of NLP models. However, existing studies mainly focus on analyzing English texts and generating adversa…