Toluwani Aremu

This paper addresses the vulnerability of existing LLM safety monitors to adaptive attackers and proposes activation watermarking, a technique that significantly improves detection robustness against such threats.

The paper argues that watermarking must be viewed as a monitoring primitive, introducing an observer-based threat model that shows even zero-bit watermarking can enable entity-level attribution through signal aggregation.

This paper demonstrates that compact, domain-specialized Automatic Speech Recognition (ASR) models significantly outperform large, general-purpose foundation models for conversational speech across 19 African languages.