cs.CRNov 16, 2026

Watermark Evidence under Open-Weight LLM Modification: Exact Bounds for Low-Rank Output Watermarks

Authors: Prashant Shekhar, Dumindu Samaraweera, Caroline Howard

Organizations: Embry-Riddle Aeronautical University, Daytona Beach, FL, USA

Abstract

Open-weight language models can continue generating after their released weights are modified, while a watermark verifier still scores outputs against the original base and marked models. We study how such modifications change the evidence available to an OpenStamp-style verifier. At each shared prefix, we represent the modification as a residual in the next-token logits. We derive an exact finite-vocabulary lower bound on the expected log-likelihood-ratio evidence under a residual-span constraint, along with a sharp span-only deviation bound. We also characterize the first-order evidence shift through a covariance term with an explicit second-order remainder, and determine the residual span needed to reach a target evidence level. For sequence-level detection, we derive an exact conditional exponential-moment bound and combine it with a supermartingale argument and a distribution-free order-statistic tolerance bound. This yields a high-confidence upper bound on miss probability after attack selection on calibration data and evaluation on independent certification continuations. In experiments with OpenStamp on Llama-2-7B and Mistral-7B-v0.3, INT8 quantization and 30% pruning retain test true-positive rates of 85.6-94.8%, while 500-step LoRA fine-tuning lowers them to 28.8% and 11.6%. The covariance prediction tracks the measured mean LoRA evidence drift on both models, with no remainder-bound violations across 160,000 test prefixes. Additionally, we observe zero false positives among 500 base-model continuations per model. Through these theoretical and empirical results, the current research provides a better understanding of watermark robustness to model-editing/tuning attacks.

Figures & tables

Explore similar work

CardsList
  1. OpenStamp: A Watermark for Open-Source Language Models

    Aug 28, 2026Miroojin Bakshi, Saksham Rastogi, Danish PruthiText WatermarkingLLM Watermarking

  2. MarkTune: Improving the Quality-Detectability Trade-off in Model-Embedded LLM Watermarking

    Dec 3, 2025Yizhou Zhao, Zhiwei Steven Wu, Adam BlockLLM WatermarkingRobust Watermarking

  3. Watermarkability: Measuring and Improving Language-Model Compatibility with Logit-Based Watermarks

    Oct 16, 2026Zewei Deng, Mohamed Seif, Liyan XieLLM Watermarking