Back to AI News
arxiv

Codebook-Based Multi-Bit Watermarking for Short LLM Outputs

Researchers propose a multi-bit LLM watermarking scheme that maps an n-bit payload to a longer u-bit codeword and embeds it during generation via token-level logit biasing. Extraction matches the received bit sequence to the nearest codeword in a precomputed secret codebook using Hamming or Levenshtein distance. Experiments report near-perfect recovery for 16-24 bit payloads on 200-token generations and strong performance on 127-token generations for base models, with weaker results on some instruction-tuned models and very short texts. A BCH-based variant avoids codebook storage but generally reduces robustness, while the Levenshtein variant improves resilience to deletions and light paraphrasing.

AI Brief

It offers a practical approach to attributing short LLM-generated texts, a setting where existing multi-bit watermarking methods often struggle.

arXiv:2611.16890v1 Announce Type: cross Abstract: Attributing short LLM-generated texts to a specific source remains challenging for existing multi-bit watermarking methods, which often require long generations or costly decoding to recover payloads reliably. We propose a codebook-based watermarking scheme that maps an $n$-bit payload to a longer $u$-bit codeword and embeds it during generation through token-level logit biasing. During extraction, the received bit sequence is matched to the nearest codeword in a precomputed secret codebook using Hamming or Levenshtein distance. Experiments show that our method achieves near-perfect recovery for 16-24 bit payloads on 200-token generations and strong performance on 127-token generations for base models. Results are weaker on some instruction-tuned models with very short texts, where low-entropy output distributions reduce embedding reliability. We also study a BCH-based variant that avoids codebook storage but generally reduces robustness. The Levenshtein variant improves resilience to deletions and light paraphrasing, while insertions and stronger rewrites remain challenging. We analyze collision probability, codeword separation, extraction time, tokenization consistency, and text quality, highlighting both the practicality and limitations of codebook-based watermarking for short-form attribution.
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!