One powerful way to safeguard AI systems is to monitor their internal computations--where signs of undesirable behavior might be more easily detectable than from models'inputs and outputs alone. However, recent studies begin to question the premise that activations provide a unique window into model behavior. In this p...
James Oldfield, S. Im, Jay Culligan et al.· 0 citations
In decentralized consumer-to-consumer (C2C) marketplaces, people list goods, negotiate with strangers, and rate one another, so trust rests on reputation. Large language model (LLM) agents now act for users, raising risks to their money, privacy, and reputation. We introduce BazaarBench, a simulated C2C marketplace and...
Ziyan Wang, Shu-Qing Shi, James Oldfield et al.· 0 citations
This work proposes $D^2-Monitor, in which a lightweight probe runs on every sample, producing a base prediction and a hesitation estimate that decides whether to escalate the sample to a more expressive probe, and exhibits the best trade-off between effectiveness and efficiency relative to 8 baselines.
Ao Liu, Yu-Peng Chen, James Oldfield et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
