# Referring Expression Grounding on RefCOCOg (val) Source: [Referring Expression Grounding on RefCOCOg (val)]() Collection index: [llms.txt]() Data updated: 2026-06-09T10:39:28.362Z Evaluation of Referring Expression Grounding performance on the validation split of the RefCOCOg dataset, reporting Accuracy at IoU threshold 0.5. 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 | Referring Expression Grounding | | Dataset (including any stated split) | RefCOCOg (val) | | Metric | Accuracy (IoU=0.5) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 (IoU=0.5)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (IoU=0.5) | Result date | Report paper | | --- | --- | --- | --- | | Visual Para-Thinker++ (3B) | 89.8 | Not recorded | 6a279e5b90966ade919bfa4c | | Visual Para-Thinker (3B) | 87.3 | Not recorded | 6a279e5b90966ade919bfa4c | | Argus-X3-8B | 86.7 | Not recorded | 68fa78f11e209131cea7b594 | | G-DINO | 86.1 | Not recorded | 68fa78f11e209131cea7b594 | | Sequential | 86.1 | Not recorded | 6a279e5b90966ade919bfa4c | | QwenVL-7B | 85.6 | Not recorded | 68fa78f11e209131cea7b594 | | Qwen2.5-VL-3B | 84.7 | Not recorded | 6a279e5b90966ade919bfa4c | | MiniGPT-7B | 84.4 | Not recorded | 68fa78f11e209131cea7b594 | | Ferret-7B | 83.9 | Not recorded | 68fa78f11e209131cea7b594 | | InternVL2-8B | 82.7 | Not recorded | 68fa78f11e209131cea7b594 | | Shikra-7B | 82.2 | Not recorded | 68fa78f11e209131cea7b594 | | MDETR | 81.6 | Not recorded | 68fa78f11e209131cea7b594 | | UniTAB | 80 | Not recorded | 68fa78f11e209131cea7b594 | | VILLA | 76.2 | Not recorded | 68fa78f11e209131cea7b594 | | UNITER | 74 | Not recorded | 68fa78f11e209131cea7b594 | | TransVG | 68.7 | Not recorded | 68fa78f11e209131cea7b594 | | OFA-L | 67.6 | Not recorded | 68fa78f11e209131cea7b594 | | MAttNet | 66.7 | Not recorded | 68fa78f11e209131cea7b594 | ### Visual Para-Thinker++ (3B) Accuracy (IoU=0.5): 89.8. Result ID: 6a27ed5cfbc373592c257245. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B"}. This record does not assert independent reproduction. ### Visual Para-Thinker (3B) Accuracy (IoU=0.5): 87.3. Result ID: 6a27ed5cfbc373592c257244. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B"}. This record does not assert independent reproduction. ### Argus-X3-8B Accuracy (IoU=0.5): 86.7. Result ID: 6995b0fa3badfaffe6b082b2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### G-DINO Accuracy (IoU=0.5): 86.1. Result ID: 6995b0fa3badfaffe6b082ab. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Sequential Accuracy (IoU=0.5): 86.1. Result ID: 6a27ed5cfbc373592c257243. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B"}. This record does not assert independent reproduction. ### QwenVL-7B Accuracy (IoU=0.5): 85.6. Result ID: 6995b0fa3badfaffe6b082b1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5-VL-3B Accuracy (IoU=0.5): 84.7. Result ID: 6a27ed5cfbc373592c257242. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B"}. This record does not assert independent reproduction. ### MiniGPT-7B Accuracy (IoU=0.5): 84.4. Result ID: 6995b0fa3badfaffe6b082af. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Ferret-7B Accuracy (IoU=0.5): 83.9. Result ID: 6995b0fa3badfaffe6b082ae. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### InternVL2-8B Accuracy (IoU=0.5): 82.7. Result ID: 6995b0fa3badfaffe6b082b0. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Shikra-7B Accuracy (IoU=0.5): 82.2. Result ID: 6995b0fa3badfaffe6b082ad. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MDETR Accuracy (IoU=0.5): 81.6. Result ID: 6995b0fa3badfaffe6b082aa. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### UniTAB Accuracy (IoU=0.5): 80. Result ID: 6995b0fa3badfaffe6b082a9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### VILLA Accuracy (IoU=0.5): 76.2. Result ID: 6995b0fa3badfaffe6b082a8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### UNITER Accuracy (IoU=0.5): 74. Result ID: 6995b0fa3badfaffe6b082a7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### TransVG Accuracy (IoU=0.5): 68.7. Result ID: 6995b0fa3badfaffe6b082a6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### OFA-L Accuracy (IoU=0.5): 67.6. Result ID: 6995b0fa3badfaffe6b082ac. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MAttNet Accuracy (IoU=0.5): 66.7. Result ID: 6995b0fa3badfaffe6b082a5. [Reporting paper]() [Method origin paper]() [Implementation]() 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.