Search papers, labs, and topics across Lattice.
This paper introduces DHMark, a novel public-key watermarking framework for LLM-generated text that enhances provenance tracing while addressing the limitations of existing methods reliant on private detection keys. By separating payload authorization from noisy textual evidence, DHMark allows for robust watermarking that withstands common text manipulations such as truncation and copy-pasting. The framework achieves a remarkable 0.967 valid rate across various edit conditions while maintaining a 0.000 acceptance rate on negative controls, demonstrating its effectiveness in real-world applications.
DHMark achieves a 96.7% valid watermark detection rate even under aggressive text edits, revolutionizing how we ensure the integrity of LLM-generated content.
Large language model (LLM) watermarking provides an important mechanism for tracing the provenance of generated text. Existing statistical watermarks are often effective and robust, but most of them rely on private detection keys, which centralizes verification and complicates public auditing. Recent public or publicly verifiable watermarking schemes improve key management, yet many of them rely on exact recovery of embedded cryptographic strings, making them fragile under token edits, truncation, copy-paste, and low-entropy generation. This paper introduces DHMark, a public-key watermarking framework for LLM-generated text. The key idea is to separate payload authorization from noisy textual evidence. An issuer signs a short registry payload bound to a public context, and the payload is expanded into many one-bit equations. During generation, a Diffie-Hellman-guided token-labeling interface assigns each candidate token a public equation vote, and the sampler softly or selectively promotes candidates whose votes agree with the authorized payload. During verification, third-party verifiers use public information to extract token votes, aggregate them into equation-level evidence, and score only signed registry records. This design avoids exact recovery of a long embedded signature and instead treats watermark detection as registry-aided statistical evidence aggregation. We formalize the public-verification setting, analyze label pseudorandomness, registry-backed soundness, and sampling distortion, and evaluate a prototype under truncation, substitution, copy-paste, wrong-context, and plain-generation attacks. In the default 32-bit configuration, DHMark maintains at least a 0.967 valid rate across eight edit conditions while yielding a 0.000 acceptance rate on three negative controls.
