Token-Level Attribution of Multi-Mode LLM Watermarks: Optimal Rule and Belief Weighting

Summary

Researchers study token-level attribution for multi-mode Gumbel-max LLM watermarks, where a document may mix spans watermarked under different secret keys with unwatermarked text. They show that the optimal rule reports the mode with the largest prior-weighted alternative density and rejects when this density exceeds a common threshold, then propose the BLADE detector, which keeps the exact Type I error and attains higher correct-label power than compared detectors on OPT-1.3B and other open-weight models.