{"ID":5438869,"CreatedAt":"2026-07-01T01:17:58.482524686Z","UpdatedAt":"2026-07-03T12:44:19.017960396Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.31602","arxiv_id":"2606.31602","title":"Robust Text Watermarking for Large Language Models via Dual Semantic Embeddings","abstract":"This work presents Dual-Embedding Watermarking (DEW), a semantic watermarking scheme for large language models (LLMs) 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 \\mbox{token and context embeddings 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, and 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.","short_abstract":"This work presents Dual-Embedding Watermarking (DEW), a semantic watermarking scheme for large language models (LLMs) 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...","url_abs":"https://arxiv.org/abs/2606.31602","url_pdf":"https://arxiv.org/pdf/2606.31602v1","authors":"[\"Jonas Schäfer\",\"Cezary Pilaszewicz\",\"Gerhard Wunder\"]","published":"2026-06-30T12:51:30Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.CR\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}