Beyond Heuristic Tuning: Power-Calibrated LLM Watermarking
For practitioners deploying LLM watermarking, this framework replaces heuristic tuning with principled hyperparameter selection to achieve optimal detectability-distortion tradeoffs.
This work develops a power-calibrated statistical framework for logit-based LLM watermarking that establishes explicit relationships between hyperparameters, detection power, and distortion, enabling optimal tradeoffs. Experiments across multiple models and datasets validate the theory and identify Pareto-optimal points.
Logit-based watermarking is a widely used mechanism for identifying LLM generated content, yet its effectiveness is governed by a fundamental trade-off between detectability and semantic distortion. Existing analyses provide limited guidance for principled hyperparameter selection, leaving practical deployments reliant on heuristic tuning. In this work, we develop a power-calibrated statistical framework that establishes explicit quantitative relationships between watermark hyperparameters, detection power, and distortion. This characterization transforms watermark design into a guided optimization problem. Building on these results, we derive practical parameter selection procedures that achieve optimal tradeoffs under constraints. Extensive experiments across multiple language models and datasets validate the theory and demonstrate that the proposed framework consistently identifies Pareto-optimal points.