SOURCE-LINKED INTELLIGENCE

Easy to Catch a Liar, Hard to Clear an Honest One: Language Models Diagnosing a Corrupted Reward Channel from a Verified Record

arXiv · AI, language, vision and robotics · article · Sep 15, 2026 · UTC

An agent that learns from rewards has to trust whatever reports those rewards. When the reports suddenly change, either the world changed or the reporter broke. From the reports alone these are indistinguishable, and reinforcement learning theory shows that no amount of further experience separates them. The prescribed escape is richer data about the reporter itself. We ask whether a frozen language model, handed exactly that data, uses it. We build a two-option game in which a payout swap and a lying reporter produce byte-identical histories. Then we add one verified record: an independent ch

Read original source ↗ Open in workspace

recordType

paper

region

Global