# Visual Grounding on LISA (test) Source: [Visual Grounding on LISA (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:31:57.736Z Evaluation of Visual Grounding performance on the test split of the LISA dataset. 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 | Visual Grounding | | Dataset (including any stated split) | LISA (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 12 of 12 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | ECVGPO | 81.14 | Not recorded | 6937acf86cfd488761a4e6ac | | GRPO | 78.78 | Not recorded | 6937acf86cfd488761a4e6ac | | GRPO | 65.9 | Not recorded | 6937acf86cfd488761a4e6ac | | ECVGPO | 65.16 | Not recorded | 6937acf86cfd488761a4e6ac | | ECVGPO | 59.26 | Not recorded | 6937acf86cfd488761a4e6ac | | SFT | 58.38 | Not recorded | 6937acf86cfd488761a4e6ac | | Base Model | 56.82 | Not recorded | 6937acf86cfd488761a4e6ac | | GRPO | 51.12 | Not recorded | 6937acf86cfd488761a4e6ac | | SFT | 48.91 | Not recorded | 6937acf86cfd488761a4e6ac | | SFT | 42.1 | Not recorded | 6937acf86cfd488761a4e6ac | | Base Model | 0.84 | Not recorded | 6937acf86cfd488761a4e6ac | | Base Model | 0.13 | Not recorded | 6937acf86cfd488761a4e6ac | ### ECVGPO Accuracy: 81.14. Result ID: 699880765698665ec4e1a156. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"InternVL3-8B-Instruct"}. This record does not assert independent reproduction. ### GRPO Accuracy: 78.78. Result ID: 699880765698665ec4e1a155. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"InternVL3-8B-Instruct"}. This record does not assert independent reproduction. ### GRPO Accuracy: 65.9. Result ID: 699880765698665ec4e1a14d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B-Instruct"}. This record does not assert independent reproduction. ### ECVGPO Accuracy: 65.16. Result ID: 699880765698665ec4e1a14e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B-Instruct"}. This record does not assert independent reproduction. ### ECVGPO Accuracy: 59.26. Result ID: 699880765698665ec4e1a152. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"InternVL3-2B-Instruct"}. This record does not assert independent reproduction. ### SFT Accuracy: 58.38. Result ID: 699880765698665ec4e1a154. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"InternVL3-8B-Instruct"}. This record does not assert independent reproduction. ### Base Model Accuracy: 56.82. Result ID: 699880765698665ec4e1a14b. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B-Instruct"}. This record does not assert independent reproduction. ### GRPO Accuracy: 51.12. Result ID: 699880765698665ec4e1a151. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"InternVL3-2B-Instruct"}. This record does not assert independent reproduction. ### SFT Accuracy: 48.91. Result ID: 699880765698665ec4e1a150. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"InternVL3-2B-Instruct"}. This record does not assert independent reproduction. ### SFT Accuracy: 42.1. Result ID: 699880765698665ec4e1a14c. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B-Instruct"}. This record does not assert independent reproduction. ### Base Model Accuracy: 0.84. Result ID: 699880765698665ec4e1a153. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"InternVL3-8B-Instruct"}. This record does not assert independent reproduction. ### Base Model Accuracy: 0.13. Result ID: 699880765698665ec4e1a14f. [Reporting paper]() Recorded evaluation configuration: {"Backbone":"InternVL3-2B-Instruct"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.