6 papers
Useless but Safe? Benchmarking Utility Recovery with User Intent Clarification in Multi-Turn Conversations
Mingqian Zheng, Malia Morgan, Liwei Jiang +2
Current LLM safety alignment techniques improve model robustness against adversarial attacks, but overlook whether and how LLMs can recover helpfulness when benign users clarify th…
Pluralistic Behavior Suite: Stress-Testing Multi-Turn Adherence to Custom Behavioral Policies
Prasoon Varshney, Makesh Narsimhan Sreedhar, Liwei Jiang +2
Large language models (LLMs) are typically aligned to a universal set of safety and usage principles intended for broad public acceptability. Yet, real-world applications of LLMs o…
Artificial Hivemind: The Open-Ended Homogeneity of Language Models (and Beyond)
Liwei Jiang, Yuanjun Chai, Margaret Li +7
Language models (LMs) often struggle to generate diverse, human-like creative content, raising concerns about the long-term homogenization of human thought through repeated exposur…
PolyGuard: A Multilingual Safety Moderation Tool for 17 Languages
Priyanshu Kumar, Devansh Jain, Akhila Yerukola +4
Truly multilingual safety moderation efforts for Large Language Models (LLMs) have been hindered by a narrow focus on a small set of languages (e.g., English, Chinese) as well as a…
SafetyAnalyst: Interpretable, Transparent, and Steerable Safety Moderation for AI Behavior
Jing-Jing Li, Valentina Pyatkin, Max Kleiman-Weiner +7
The ideal AI safety moderation system would be both structurally interpretable (so its decisions can be reliably explained) and steerable (to align to safety standards and reflect…
To Err is AI : A Case Study Informing LLM Flaw Reporting Practices
Sean McGregor, Allyson Ettinger, Nick Judd +10
In August of 2024, 495 hackers generated evaluations in an open-ended bug bounty targeting the Open Language Model (OLMo) from The Allen Institute for AI. A vendor panel staffed by…