Refined Detection for Gumbel Watermarking
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910090374676480 |
|---|---|
| author | Lattimore, Tor |
| author_facet | Lattimore, Tor |
| contents | We propose a simple detection mechanism for the Gumbel watermarking scheme proposed by Aaronson (2022). The new mechanism is proven to be near-optimal in a problem-dependent sense among all model-agnostic watermarking schemes under the assumption that the next-token distribution is sampled i.i.d. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_30017 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Refined Detection for Gumbel Watermarking Lattimore, Tor Machine Learning Cryptography and Security We propose a simple detection mechanism for the Gumbel watermarking scheme proposed by Aaronson (2022). The new mechanism is proven to be near-optimal in a problem-dependent sense among all model-agnostic watermarking schemes under the assumption that the next-token distribution is sampled i.i.d. |
| title | Refined Detection for Gumbel Watermarking |
| topic | Machine Learning Cryptography and Security |
| url | https://arxiv.org/abs/2603.30017 |