# Watermark Detectability on C4-RealNewsLike Dipper-1 (test) Source: [Watermark Detectability on C4-RealNewsLike Dipper-1 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:25:07.546Z Evaluation of watermark detectability performance using AUC and TPR on the test split of the C4-RealNewsLike dataset subjected to the Dipper-1 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 Dipper-1 (test) | | Metric | AUC | | Unit | SCALAR | | 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: SCALAR. - [TPR](): Higher is better; unit: SCALAR. ## Recorded results | Method | AUC | Result date | Report paper | | --- | --- | --- | --- | | MarkTune | 0.977 | Not recorded | 693115d9d6b0dd3342933da6 | | MarkTune | 0.961 | Not recorded | 693115d9d6b0dd3342933da6 | | Gumbel-max | 0.95 | Not recorded | 693115d9d6b0dd3342933da6 | | Gumbel-max | 0.945 | Not recorded | 693115d9d6b0dd3342933da6 | | SynthID | 0.943 | Not recorded | 693115d9d6b0dd3342933da6 | | SynthID | 0.941 | Not recorded | 693115d9d6b0dd3342933da6 | | GaussMark | 0.915 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-2 | 0.884 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-2 | 0.862 | Not recorded | 693115d9d6b0dd3342933da6 | | GaussMark | 0.857 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-D | 0.855 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-D | 0.814 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-1 | 0.771 | Not recorded | 693115d9d6b0dd3342933da6 | | KGW-1 | 0.757 | Not recorded | 693115d9d6b0dd3342933da6 | ### MarkTune AUC: 0.977. Result ID: 69998b265698665ec4fd0d43. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### MarkTune AUC: 0.961. Result ID: 69998b265698665ec4fd0d4a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### Gumbel-max AUC: 0.95. Result ID: 69998b265698665ec4fd0d46. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### Gumbel-max AUC: 0.945. Result ID: 69998b265698665ec4fd0d3f. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### SynthID AUC: 0.943. Result ID: 69998b265698665ec4fd0d47. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### SynthID AUC: 0.941. Result ID: 69998b265698665ec4fd0d40. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### GaussMark AUC: 0.915. Result ID: 69998b265698665ec4fd0d42. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-2 AUC: 0.884. Result ID: 69998b265698665ec4fd0d3e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-2 AUC: 0.862. Result ID: 69998b265698665ec4fd0d45. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### GaussMark AUC: 0.857. Result ID: 69998b265698665ec4fd0d49. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-D AUC: 0.855. Result ID: 69998b265698665ec4fd0d41. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-D AUC: 0.814. Result ID: 69998b265698665ec4fd0d48. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Llama2-7B","Temperature":"1.0"}. This record does not assert independent reproduction. ### KGW-1 AUC: 0.771. Result ID: 69998b265698665ec4fd0d3d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-4B","Temperature":"0.7"}. This record does not assert independent reproduction. ### KGW-1 AUC: 0.757. Result ID: 69998b265698665ec4fd0d44. [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.