[2601.22246] MirrorMark: A Distortion-Free Multi-Bit Watermark for Large Language Models

Computer Science > Cryptography and Security arXiv:2601.22246 (cs) [Submitted on 29 Jan 2026 (v1), last revised 27 Apr 2026 (this version, v2)] Title:MirrorMark: A Distortion-Free Multi-Bit Watermark for Large Language Models Authors:Ya Jiang, Massieh Kordi Boroujeny, Surender Suresh Kumar, Kai Zeng View a PDF of the paper titled MirrorMark: A Distortion-Free Multi-Bit Watermark for Large Language Models, by Ya Jiang and 3 other authors View PDF HTML (experimental) Abstract:As large language models (LLMs) become integral to applications such as question answering and content creation, reliable content attribution has become increasingly important. Watermarking is a promising approach, but existing methods either provide only binary signals or distort the sampling distribution, degrading text quality; distortion-free approaches, in turn, often suffer from weak detectability or robustness. We propose MirrorMark, a multi-bit and distortion-free watermark for LLMs. By mirroring sampling randomness in a measure-preserving manner, MirrorMark embeds multi-bit messages without altering the token probability distribution, preserving text quality by design. To improve robustness, we introduce a context-based scheduler that balances token assignments across message positions while remaining resilient to insertions and deletions. We further provide a theoretical analysis of the equal error rate to interpret empirical performance. Experiments show that MirrorMark matches the text quali...