# SWAN: Semantic Watermarking with Abstract Meaning Representation

arXiv: https://arxiv.org/abs/2605.04305 · HTML: https://mlsanity.com/paper/2605.04305/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 26 references that are in the map.
- First posted on arXiv 5 May 2026.

## Builds on (most cited first, 12 of 12)

- [PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization](https://mlsanity.com/paper/1912.08777/) (2019)
- [Robust Distortion-free Watermarks for Language Models](https://mlsanity.com/paper/2307.15593/) (2023)
- [Exploring the Use of Large Language Models for Reference-Free Text Quality Evaluation: An Empirical Study](https://mlsanity.com/paper/2304.00723/) (2023)
- [Robust Multi-bit Natural Language Watermarking through Invariant Features](https://mlsanity.com/paper/2305.01904/) (2023)
- [A Robust Semantics-based Watermark for Large Language Model against Paraphrasing](https://mlsanity.com/paper/2311.08721/) (2023)
- [k-SemStamp: A Clustering-Based Semantic Watermark for Detection of Machine-Generated Text](https://mlsanity.com/paper/2402.11399/) (2024)
- [Watermarking Pre-trained Language Models with Backdooring](https://mlsanity.com/paper/2210.07543/) (2022)
- [AMR Similarity Metrics from Principles](https://mlsanity.com/paper/2001.10929/) (2020)
- [Large language models can consistently generate high-quality content for election disinformation operations](https://mlsanity.com/paper/2408.06731/) (2024)
- [PART: Pre-trained Authorship Representation Transformer](https://mlsanity.com/paper/2209.15373/) (2022)
- [Do LLMs Know to Respect Copyright Notice?](https://mlsanity.com/paper/2411.01136/) (2024)
- [MASSIVE Multilingual Abstract Meaning Representation: A Dataset and Baselines for Hallucination Detection](https://mlsanity.com/paper/2405.19285/) (2024)

## Abstract (arXiv)

We introduce SWAN (Semantic Watermarking with Abstract Meaning Representation), a novel framework that embeds watermark signatures into the semantic structure of a sentence using Abstract Meaning Representation (AMR). In contrast to existing watermarking methods, which typically encode signatures by adjusting token selection preferences during text generation, SWAN embeds the signature directly in the sentence's semantic representation. As the signature is encoded at the semantic structure level, any paraphrase that preserves meaning automatically preserves the signature. SWAN is training-free: watermark injection is achieved by prompting an LLM to generate sentences guided by a selected AMR template while maintaining contextual coherence, and detection uses an off-the-shelf AMR parser followed by a simple one-proportion z-test. Empirical evaluation on the RealNews benchmark shows SWAN matches state-of-the-art detection performance on unaltered watermarked text, while significantly improving robustness against paraphrasing, increasing detection AUC by up to 13.9 percentage points compared to prior methods. These results demonstrate that SWAN's approach of anchoring watermarks in AMR semantic structures provides a simple, effective, and prompt-based method for robust text provenance verification under paraphrasing, opening new avenues for semantic-level watermarking research.