# Distortion-Free Watermark Evaluation on ELI5 16-bit bitstrings LLAMA3.1-8B Source: [Distortion-Free Watermark Evaluation on ELI5 16-bit bitstrings LLAMA3.1-8B]() Collection index: [llms.txt]() Data updated: 2026-05-13T09:14:07.708Z Evaluation of Distortion-Free Watermark performance on text generated by the LLaMA 3.1 8B model using the ELI5 16-bit bitstrings dataset. Key metrics include Message Accuracy and Bit Accuracy at 1% False Positive Rate. 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 | Distortion-Free Watermark Evaluation | | Dataset (including any stated split) | ELI5 16-bit bitstrings LLAMA3.1-8B | | Metric | Message Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 8 of 8 | | 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 - [Message Accuracy](): Higher is better; unit: PERCENT. - [Bit Accuracy (1% FPR)](): Higher is better; unit: PERCENT. - [Bit Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Message Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Cycle-Shift | 77.8 | Not recorded | 6a0405c4dbc836b87d8277a1 | | ArcMark | 41.7 | Not recorded | 6a0405c4dbc836b87d8277a1 | | Ours (SynthID) | 15.8 | Not recorded | 6a0405c4dbc836b87d8277a1 | | BiMark | 11.4 | Not recorded | 6a0405c4dbc836b87d8277a1 | | MC2Mark | 9.7 | Not recorded | 6a0405c4dbc836b87d8277a1 | | MirrorMark | 5.7 | Not recorded | 6a0405c4dbc836b87d8277a1 | | Ours+ (SynthID) | 4.5 | Not recorded | 6a0405c4dbc836b87d8277a1 | | StealthInk | 1.1 | Not recorded | 6a0405c4dbc836b87d8277a1 | ### Cycle-Shift Message Accuracy: 77.8. Result ID: 6a04400d47f34ce924ea53cb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ArcMark Message Accuracy: 41.7. Result ID: 6a04400d47f34ce924ea53cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Ours (SynthID) Message Accuracy: 15.8. Result ID: 6a04400d47f34ce924ea53d1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"n\_layer":"100","Encoder":"stateless"}. This record does not assert independent reproduction. ### BiMark Message Accuracy: 11.4. Result ID: 6a04400d47f34ce924ea53cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MC2Mark Message Accuracy: 9.7. Result ID: 6a04400d47f34ce924ea53ce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MirrorMark Message Accuracy: 5.7. Result ID: 6a04400d47f34ce924ea53cf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Ours+ (SynthID) Message Accuracy: 4.5. Result ID: 6a04400d47f34ce924ea53d2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"n\_layer":"100","Encoder":"stateful"}. This record does not assert independent reproduction. ### StealthInk Message Accuracy: 1.1. Result ID: 6a04400d47f34ce924ea53d0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.