# Referring Expression Comprehension on RefCOCOg (val-g) Source: [Referring Expression Comprehension on RefCOCOg (val-g)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:30:05.357Z Evaluation of Referring Expression Comprehension performance on the val-g split of the RefCOCOg dataset, typically measured using Accuracy. 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 Comprehension | | Dataset (including any stated split) | RefCOCOg (val-g) | | Metric | Accuracy | | Unit | PERCENT | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | SimVG-DB | 80.42 | Not recorded | 68fa756f1e209131cea67e78 | | SimVG-TB | 79.34 | Not recorded | 68fa756f1e209131cea67e78 | | SimVG-DB | 78.81 | Not recorded | 68fa756f1e209131cea67e78 | | SimVG-TB | 77.66 | Not recorded | 68fa756f1e209131cea67e78 | | TransCP | 72.6 | Not recorded | 68fa756f1e209131cea67e78 | | VLTVG | 72.53 | Not recorded | 68fa756f1e209131cea67e78 | | Dyn.MDETR | 72.21 | Not recorded | 68fa756f1e209131cea67e78 | | SeqTR | 71.5 | Not recorded | 68fa756f1e209131cea67e78 | | DGA | 69.07 | Not recorded | 68fa756f1e209131cea67e78 | | TransVG | 67.02 | Not recorded | 68fa756f1e209131cea67e78 | | ReSCL | 63.12 | Not recorded | 68fa756f1e209131cea67e78 | | LBYL | 62.7 | Not recorded | 68fa756f1e209131cea67e78 | | VGTR | 61.64 | Not recorded | 68fa756f1e209131cea67e78 | | NMTree | 57.52 | Not recorded | 68fa756f1e209131cea67e78 | ### SimVG-DB Accuracy: 80.42. Result ID: 6993be213a6c3148491836c2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Visual Encoder":"ViT-L/32","Time (ms)":"116"}. This record does not assert independent reproduction. ### SimVG-TB Accuracy: 79.34. Result ID: 6993be213a6c3148491836c1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Visual Encoder":"ViT-L/32","Time (ms)":"101"}. This record does not assert independent reproduction. ### SimVG-DB Accuracy: 78.81. Result ID: 6993be213a6c3148491836c0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Visual Encoder":"ViT-B/32","Time (ms)":"52"}. This record does not assert independent reproduction. ### SimVG-TB Accuracy: 77.66. Result ID: 6993be213a6c3148491836bf. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Visual Encoder":"ViT-B/32","Time (ms)":"44"}. This record does not assert independent reproduction. ### TransCP Accuracy: 72.6. Result ID: 6993be213a6c3148491836bd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"RN50","Time (ms)":"74\*"}. This record does not assert independent reproduction. ### VLTVG Accuracy: 72.53. Result ID: 6993be213a6c3148491836bc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"RN50","Time (ms)":"79\*"}. This record does not assert independent reproduction. ### Dyn.MDETR Accuracy: 72.21. Result ID: 6993be213a6c3148491836be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"ViT-B/16"}. This record does not assert independent reproduction. ### SeqTR Accuracy: 71.5. Result ID: 6993be213a6c3148491836bb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Visual Encoder":"DN53","Time (ms)":"50"}. This record does not assert independent reproduction. ### DGA Accuracy: 69.07. Result ID: 6993be213a6c3148491836b5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"VGG16","Time (ms)":"341"}. This record does not assert independent reproduction. ### TransVG Accuracy: 67.02. Result ID: 6993be213a6c3148491836b9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"RN101","Time (ms)":"62"}. This record does not assert independent reproduction. ### ReSCL Accuracy: 63.12. Result ID: 6993be213a6c3148491836b7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"DN53","Time (ms)":"36"}. This record does not assert independent reproduction. ### LBYL Accuracy: 62.7. Result ID: 6993be213a6c3148491836b8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Visual Encoder":"DN53","Time (ms)":"30"}. This record does not assert independent reproduction. ### VGTR Accuracy: 61.64. Result ID: 6993be213a6c3148491836ba. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"RN50"}. This record does not assert independent reproduction. ### NMTree Accuracy: 57.52. Result ID: 6993be213a6c3148491836b6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"RN101"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.