# Watermark Detectability on C4 RealNewsLike (Del-0.2) (test) Source: [Watermark Detectability on C4 RealNewsLike (Del-0.2) (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:25:07.318Z Evaluation of watermark detectability performance on the specific test split of the C4 RealNewsLike dataset subjected to the Del-0.2 perturbation, reporting metrics such as AUC and 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 (Del-0.2) (test) | | Metric | AUC | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 28 of 28 | | 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 | | --- | --- | --- | --- | | Gumbel-max | 99.3 | Not recorded | 693115d9d6b0dd3342933da6 | | Gumbel-max | 99.3 | Not recorded | 693115d9d6b0dd3342933da6 | | Gumbel-max | 99.3 | Not recorded | 693115d9d6b0dd3342933da6 | | Gumbel-max | 99.3 | Not recorded | 693115d9d6b0dd3342933da6 | | SynthID | 98.7 | Not recorded | 693115d9d6b0dd3342933da6 | | SynthID | 98.7 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-2 | 98.6 | Not recorded | 693115d9d6b0dd3342933da6 | | SynthID | 98.6 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-2 | 98 | Not recorded | 693115d9d6b0dd3342933da6 | | SynthID | 97.9 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-2 | 97.5 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-2 | 97.5 | Not recorded | 693115d9d6b0dd3342933da6 | | MarkTune | 94.8 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-D | 94.7 | Not recorded | 693115d9d6b0dd3342933da6 | | MarkTune | 94.4 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-D | 93.9 | Not recorded | 693115d9d6b0dd3342933da6 | | MarkTune | 92.8 | Not recorded | 693115d9d6b0dd3342933da6 | | MarkTune | 92.1 | Not recorded | 693115d9d6b0dd3342933da6 | | GaussMark | 91.9 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-D | 91.4 | Not recorded | 693115d9d6b0dd3342933da6 | | GaussMark | 90.8 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-D | 88.7 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-1 | 88.1 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-1 | 87.8 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-1 | 87.3 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-1 | 85.3 | Not recorded | 693115d9d6b0dd3342933da6 | | GaussMark | 85.3 | Not recorded | 693115d9d6b0dd3342933da6 | | GaussMark | 83.5 | Not recorded | 693115d9d6b0dd3342933da6 | ### Gumbel-max AUC: 99.3. Result ID: 69998b265698665ec4fd0d07. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### Gumbel-max AUC: 99.3. Result ID: 69998b265698665ec4fd0d0e. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### Gumbel-max AUC: 99.3. Result ID: 69998b265698665ec4fd0d23. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### Gumbel-max AUC: 99.3. Result ID: 69998b265698665ec4fd0d2a. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### SynthID AUC: 98.7. Result ID: 69998b265698665ec4fd0d08. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### SynthID AUC: 98.7. Result ID: 69998b265698665ec4fd0d2b. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-2 AUC: 98.6. Result ID: 69998b265698665ec4fd0d22. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### SynthID AUC: 98.6. Result ID: 69998b265698665ec4fd0d24. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-2 AUC: 98. Result ID: 69998b265698665ec4fd0d29. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### SynthID AUC: 97.9. Result ID: 69998b265698665ec4fd0d0f. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-2 AUC: 97.5. Result ID: 69998b265698665ec4fd0d06. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-2 AUC: 97.5. Result ID: 69998b265698665ec4fd0d0d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### MarkTune AUC: 94.8. Result ID: 69998b265698665ec4fd0d27. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-D AUC: 94.7. Result ID: 69998b265698665ec4fd0d09. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### MarkTune AUC: 94.4. Result ID: 69998b265698665ec4fd0d2e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-D AUC: 93.9. Result ID: 69998b265698665ec4fd0d25. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### MarkTune AUC: 92.8. Result ID: 69998b265698665ec4fd0d12. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### MarkTune AUC: 92.1. Result ID: 69998b265698665ec4fd0d0b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### GaussMark AUC: 91.9. Result ID: 69998b265698665ec4fd0d26. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-D AUC: 91.4. Result ID: 69998b265698665ec4fd0d10. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### GaussMark AUC: 90.8. Result ID: 69998b265698665ec4fd0d0a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-D AUC: 88.7. Result ID: 69998b265698665ec4fd0d2c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-1 AUC: 88.1. Result ID: 69998b265698665ec4fd0d05. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-1 AUC: 87.8. Result ID: 69998b265698665ec4fd0d28. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-1 AUC: 87.3. Result ID: 69998b265698665ec4fd0d21. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-1 AUC: 85.3. Result ID: 69998b265698665ec4fd0d0c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### GaussMark AUC: 85.3. Result ID: 69998b265698665ec4fd0d2d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### GaussMark AUC: 83.5. Result ID: 69998b265698665ec4fd0d11. [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.