The Pushback Paradox: A Two-Probe Diagnostic for Language Model Compliance
Are language models compliant with user instructions? A model that always complies can be stopped but also exploited, while one that always resists can be neither exploited nor stopped. We contribute an open two-probe benchmark that can place any language model on this spectrum. In the active probe, a user instructs th...
Stefan Bühler, David Exler, M. Reischl et al.
· 0 citations