# Watermark Detectability on C4-RealNewsLike Translate (test) Source: [Watermark Detectability on C4-RealNewsLike Translate (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:25:07.590Z Evaluation of Watermark Detectability performance on the test split of the C4-RealNewsLike dataset subjected to Translation perturbations, reporting AUC and True Positive Rate (TPR). A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Watermark Detectability | | Dataset (including any stated split) | C4-RealNewsLike Translate (test) | | Metric | AUC | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 14 of 14 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [AUC](): Higher is better; unit: PERCENT. - [TPR](): Higher is better; unit: PERCENT. ## Recorded results | Method | AUC | Result date | Report paper | | --- | --- | --- | --- | | MarkTune | 97.3 | Not recorded | 693115d9d6b0dd3342933da6 | | SynthID | 96.8 | Not recorded | 693115d9d6b0dd3342933da6 | | MarkTune | 96.6 | Not recorded | 693115d9d6b0dd3342933da6 | | Gumbel-max | 96.1 | Not recorded | 693115d9d6b0dd3342933da6 | | Gumbel-max | 95.5 | Not recorded | 693115d9d6b0dd3342933da6 | | SynthID | 94.9 | Not recorded | 693115d9d6b0dd3342933da6 | | GaussMark | 94.8 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-2 | 92.8 | Not recorded | 693115d9d6b0dd3342933da6 | | GaussMark | 92.5 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-2 | 92.2 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-D | 91.2 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-D | 88.9 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-1 | 83.1 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-1 | 77.7 | Not recorded | 693115d9d6b0dd3342933da6 | ### MarkTune AUC: 97.3. Result ID: 69998b265698665ec4fd0d5f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### SynthID AUC: 96.8. Result ID: 69998b265698665ec4fd0d5c. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### MarkTune AUC: 96.6. Result ID: 69998b265698665ec4fd0d66. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### Gumbel-max AUC: 96.1. Result ID: 69998b265698665ec4fd0d62. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### Gumbel-max AUC: 95.5. Result ID: 69998b265698665ec4fd0d5b. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### SynthID AUC: 94.9. Result ID: 69998b265698665ec4fd0d63. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### GaussMark AUC: 94.8. Result ID: 69998b265698665ec4fd0d5e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-2 AUC: 92.8. Result ID: 69998b265698665ec4fd0d5a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### GaussMark AUC: 92.5. Result ID: 69998b265698665ec4fd0d65. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-2 AUC: 92.2. Result ID: 69998b265698665ec4fd0d61. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-D AUC: 91.2. Result ID: 69998b265698665ec4fd0d5d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-D AUC: 88.9. Result ID: 69998b265698665ec4fd0d64. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-1 AUC: 83.1. Result ID: 69998b265698665ec4fd0d59. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-1 AUC: 77.7. Result ID: 69998b265698665ec4fd0d60. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.