SAMark: A Self-Anchored Text Watermarking with Paragraph-Level Paraphrase Robustness
About
Semantic-level watermarking (SWM) improves robustness against text modifications by treating sentences as the basic unit. However, robustness to paragraph-level paraphrasing remains difficult because such attacks globally disrupt watermark signals by changing sentence order. In this work, we propose SAMark, a self-anchored watermarking framework that removes the dependency on sentence order by establishing a step-independent green region in semantic space. To improve detectability, we introduce a multi-channel hyperbolic scoring mechanism that amplifies watermark signals while suppressing noise from weakly aligned candidates. We further propose a diversity-aware filtering strategy that combines hard filtering with soft regularization, extending beyond simple n-gram repetition filters to address semantic redundancy. Experimental results show that SAMark achieves up to 90.2% TP@FP1% under typical paragraph-level paraphrasing attacks, outperforming the strongest prior baseline by more than 30% on average, while maintaining generation quality competitive with unwatermarked text and breaking the robustness-quality trade-off that limits prior methods.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Large Language Model Watermarking | BookSum (test) | Average Rank5.14 | 20 | |
| Watermark Detection | C4 | TPR @ 1% FPR (No Attack)95.2 | 20 | |
| Text Quality Assessment | C4 | Average Rank5.98 | 20 | |
| Watermark Detection | BOOKSUM Mistral-Small-3.1-24B-Base-2503 (test) | Latency per Token0.0014 | 9 | |
| Watermarked Text Generation | BOOKSUM Mistral-Small-3.1-24B-Base-2503 (test) | Latency per Token (s/tok)0.218 | 9 |