Mixture-of-Experts (MoE) language models produce routing information during inference that may be logged or exposed for monitoring, debugging, load analysis, and safety auditing. Unlike ordinary model outputs, this telemetry reveals a view of the model's internal computation, raising a privacy question: can it reveal w …
When a prompt injection attack succeeds, a Large Language Model (LLM) abandons its assigned system role to comply with an adversarial instruction. While prior work has extensively quantified how often this occurs, we ask a more fundamental question: where inside the network does the model actually decide to break the r …
Membership inference (MI) asks whether a specific record appeared in a model's training set and is increasingly used as evidence for data provenance and copyright auditing. These applications require determining whether the exact queried record was used for training, rather than merely whether the model was exposed to …