# Multimodal Watermarking on MS-COCO 17 Source: [Multimodal Watermarking on MS-COCO 17]() Collection index: [llms.txt]() Data updated: 2026-02-27T11:24:26.000Z Evaluation of Multimodal Watermarking performance on the MS-COCO 2017 dataset, reporting metrics such as AUC, PPL, BertScore, and Chair-I. 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 | Multimodal Watermarking | | Dataset (including any stated split) | MS-COCO 17 | | Metric | PPL | | Unit | SCALAR | | Direction | Lower 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 - [PPL](): Lower is better; unit: SCALAR. - [AUC](): Higher is better; unit: SCALAR. - [BertScore](): Higher is better; unit: SCALAR. - [Chair-I](): Higher is better; unit: SCALAR. ## Recorded results | Method | PPL | Result date | Report paper | | --- | --- | --- | --- | | NW | 3.01 | Not recorded | 6965d19a6343fa78c8323688 | | VISA-M | 3.02 | Not recorded | 6965d19a6343fa78c8323688 | | VLA | 3.08 | Not recorded | 6965d19a6343fa78c8323688 | | KGW | 3.11 | Not recorded | 6965d19a6343fa78c8323688 | | DiP | 3.13 | Not recorded | 6965d19a6343fa78c8323688 | | Unbiased | 3.14 | Not recorded | 6965d19a6343fa78c8323688 | | SWEET | 3.18 | Not recorded | 6965d19a6343fa78c8323688 | | NW | 5.23 | Not recorded | 6965d19a6343fa78c8323688 | | VISA-M | 5.59 | Not recorded | 6965d19a6343fa78c8323688 | | SWEET | 5.69 | Not recorded | 6965d19a6343fa78c8323688 | | DiP | 5.7 | Not recorded | 6965d19a6343fa78c8323688 | | Unbiased | 5.7 | Not recorded | 6965d19a6343fa78c8323688 | | KGW | 5.79 | Not recorded | 6965d19a6343fa78c8323688 | | VLA | 5.81 | Not recorded | 6965d19a6343fa78c8323688 | ### NW PPL: 3.01. Result ID: 69a17edd35dfcabb00e76e9c. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### VISA-M PPL: 3.02. Result ID: 69a17edd35dfcabb00e76ea2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### VLA PPL: 3.08. Result ID: 69a17edd35dfcabb00e76e9d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### KGW PPL: 3.11. Result ID: 69a17edd35dfcabb00e76e9e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### DiP PPL: 3.13. Result ID: 69a17edd35dfcabb00e76ea0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### Unbiased PPL: 3.14. Result ID: 69a17edd35dfcabb00e76ea1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### SWEET PPL: 3.18. Result ID: 69a17edd35dfcabb00e76e9f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen"}. This record does not assert independent reproduction. ### NW PPL: 5.23. Result ID: 69a17edd35dfcabb00e76e95. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaVA"}. This record does not assert independent reproduction. ### VISA-M PPL: 5.59. Result ID: 69a17edd35dfcabb00e76e9b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA"}. This record does not assert independent reproduction. ### SWEET PPL: 5.69. Result ID: 69a17edd35dfcabb00e76e98. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaVA"}. This record does not assert independent reproduction. ### DiP PPL: 5.7. Result ID: 69a17edd35dfcabb00e76e99. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA"}. This record does not assert independent reproduction. ### Unbiased PPL: 5.7. Result ID: 69a17edd35dfcabb00e76e9a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA"}. This record does not assert independent reproduction. ### KGW PPL: 5.79. Result ID: 69a17edd35dfcabb00e76e97. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA"}. This record does not assert independent reproduction. ### VLA PPL: 5.81. Result ID: 69a17edd35dfcabb00e76e96. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.