Reducing False Positives in Watermark Detection through Low-Entropy Token Filtering

Abstract

This thesis addresses the problem of false positives in watermark detection for large language models (LLMs), with a particular focus on open-domain news-style text. As LLMs increasingly generate content that mimics human writing, the risk of mistakenly classifying human-written text as watermarked poses serious ethical and practical concerns. To mitigate this issue, this study investigates two strategies that extend the KGW framework: named entity-based weighting and entropy-based filtering. The proposed methods aim to reduce the inf Show more