This work presents Dual-Embedding Watermarking (DEW), a semantic watermarking scheme for large language models that leverages contextual and token-level embeddings to enhance robustness against paraphrasing and translation. DEW utilizes a signal-processing methodology, applying algebraic vector-space operations to derive a watermark signal that degrades gracefully under semantic shifts.
- The method obfuscates the watermark by projecting embedding vectors through pseudo-random matrices seeded with a secret key.
- Relevant distributions derived from the underlying algebra are evaluated and employed for statistical testing and benchmarking of DEW.
- Experimental results across multiple LLMs indicate that DEW improves post-paraphrase detection while maintaining competitive text quality.
- The watermark remains detectable after translation, even when prior semantic watermarks degrade significantly.
These findings position DEW as a practical and robust solution for safeguarding LLM-generated text and addressing critical issues in responsible AI deployment.