Alignment tuning has become the de facto standard practice for enabling base large
language models (LLMs) to serve as open-domain AI assistants. The alignment tuning …

Most traditional AI safety research has approached AI models as machines and centered on
algorithm-focused attacks developed by security experts. As large language models (LLMs) …

AI safety training and red-teaming of large language models (LLMs) are measures to
mitigate the generation of unsafe content. Our work exposes the inherent cross-lingual …

Large language models (LLMs) are popular for high-quality text generation but can produce
harmful content, even when aligned with human values through reinforcement learning …

Warning: This paper contains examples of harmful language, and reader discretion is
recommended. The increasing open release of powerful large language models (LLMs) has …

Automated red teaming holds substantial promise for uncovering and mitigating the risks
associated with the malicious use of large language models (LLMs), yet the field lacks a …

Large language models (LLMs) show inherent brittleness in their safety mechanisms, as
evidenced by their susceptibility to jailbreaking and even non-malicious fine-tuning. This …

Large Language Models (LLMs) continue to advance in their capabilities, yet this progress is
accompanied by a growing array of safety risks. While significant attention has been …

Open-Domain Question Answering (ODQA) aims at answering factoid questions without
explicitly providing specific background documents. In a zero-shot setting, this task is more …

As large language models (LLMs) become increasingly integrated into real-world
applications such as code generation and chatbot assistance, extensive efforts have been …