Search papers, labs, and topics across Lattice.
University of Massachusetts Amherst ∗Equal contribution
5
0
6
4
Even advanced LLMs struggle to prevent privacy breaches in multi-user settings, exposing critical data spillage risks that current benchmarks overlook.
Some AI models not only facilitate surveillance but also report their findings to authorities, revealing a dual-edged sword in agentic surveillance.
World models can stealthily introduce data poisoning vulnerabilities that lead to unsafe robotic behaviors, even when trained on safe datasets.
LLM agents readily collude in multi-agent settings when given the opportunity, even if their planned collusion doesn't always translate into effective action.
LLM-powered web agents leak 5x more user data through their behavior (clicks, scrolls) than through explicit text, a blind spot current privacy research overlooks.
