# Watermark Detectability on C4 RealNewsLike (Del-0.5) (test) Source: [Watermark Detectability on C4 RealNewsLike (Del-0.5) (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:25:07.337Z Evaluation of watermark detectability performance using AUC and True Positive Rate (TPR) on the specific test split of the C4 RealNewsLike dataset modified by a 0.5 deletion perturbation. 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 (Del-0.5) (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 | | --- | --- | --- | --- | | KGW-D | 85.1 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-D | 80.3 | Not recorded | 693115d9d6b0dd3342933da6 | | Gumbel-max | 79 | Not recorded | 693115d9d6b0dd3342933da6 | | MarkTune | 78.3 | Not recorded | 693115d9d6b0dd3342933da6 | | SynthID | 78 | Not recorded | 693115d9d6b0dd3342933da6 | | Gumbel-max | 77.8 | Not recorded | 693115d9d6b0dd3342933da6 | | SynthID | 77.7 | Not recorded | 693115d9d6b0dd3342933da6 | | MarkTune | 74.4 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-2 | 71.1 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-2 | 69.7 | Not recorded | 693115d9d6b0dd3342933da6 | | GaussMark | 62.2 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-1 | 61.7 | Not recorded | 693115d9d6b0dd3342933da6 | | GaussMark | 61.7 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-1 | 58.1 | Not recorded | 693115d9d6b0dd3342933da6 | ### KGW-D AUC: 85.1. Result ID: 69998b265698665ec4fd0d17. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-D AUC: 80.3. Result ID: 69998b265698665ec4fd0d1e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### Gumbel-max AUC: 79. Result ID: 69998b265698665ec4fd0d1c. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### MarkTune AUC: 78.3. Result ID: 69998b265698665ec4fd0d20. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### SynthID AUC: 78. Result ID: 69998b265698665ec4fd0d16. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### Gumbel-max AUC: 77.8. Result ID: 69998b265698665ec4fd0d15. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### SynthID AUC: 77.7. Result ID: 69998b265698665ec4fd0d1d. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### MarkTune AUC: 74.4. Result ID: 69998b265698665ec4fd0d19. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-2 AUC: 71.1. Result ID: 69998b265698665ec4fd0d1b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-2 AUC: 69.7. Result ID: 69998b265698665ec4fd0d14. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### GaussMark AUC: 62.2. Result ID: 69998b265698665ec4fd0d18. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-1 AUC: 61.7. Result ID: 69998b265698665ec4fd0d13. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### GaussMark AUC: 61.7. Result ID: 69998b265698665ec4fd0d1f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-1 AUC: 58.1. Result ID: 69998b265698665ec4fd0d1a. [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.