Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911587658366976 |
|---|---|
| author | Wang, Zhuoshang Ren, Yubing Cao, Yanan Fang, Fang Li, Xiaoxue Guo, Li |
| author_facet | Wang, Zhuoshang Ren, Yubing Cao, Yanan Fang, Fang Li, Xiaoxue Guo, Li |
| contents | While watermarking serves as a critical mechanism for LLM provenance, existing secret-key schemes tightly couple detection with injection, requiring access to keys or provider-side scheme-specific detectors for verification. This dependency creates a fundamental barrier for real-world governance, as independent auditing becomes impossible without compromising model security or relying on the opaque claims of service providers. To resolve this dilemma, we introduce TTP-Detect, a pioneering black-box framework designed for non-intrusive, third-party watermark verification. By decoupling detection from injection, TTP-Detect reframes verification as a relative hypothesis testing problem. It employs a proxy model to amplify watermark-relevant signals and a suite of complementary relative measurements to assess the alignment of the query text with watermarked distributions. Extensive experiments across representative watermarking schemes, datasets and models demonstrate that TTP-Detect achieves superior detection performance and robustness against diverse attacks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_14968 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework Wang, Zhuoshang Ren, Yubing Cao, Yanan Fang, Fang Li, Xiaoxue Guo, Li Cryptography and Security Computation and Language While watermarking serves as a critical mechanism for LLM provenance, existing secret-key schemes tightly couple detection with injection, requiring access to keys or provider-side scheme-specific detectors for verification. This dependency creates a fundamental barrier for real-world governance, as independent auditing becomes impossible without compromising model security or relying on the opaque claims of service providers. To resolve this dilemma, we introduce TTP-Detect, a pioneering black-box framework designed for non-intrusive, third-party watermark verification. By decoupling detection from injection, TTP-Detect reframes verification as a relative hypothesis testing problem. It employs a proxy model to amplify watermark-relevant signals and a suite of complementary relative measurements to assess the alignment of the query text with watermarked distributions. Extensive experiments across representative watermarking schemes, datasets and models demonstrate that TTP-Detect achieves superior detection performance and robustness against diverse attacks. |
| title | Rethinking LLM Watermark Detection in Black-Box Settings: A Non-Intrusive Third-Party Framework |
| topic | Cryptography and Security Computation and Language |
| url | https://arxiv.org/abs/2603.14968 |