# Visual Grounding on ReferitGame (test) using Pr@0.5 Source: [Visual Grounding on ReferitGame (test) using Pr@0.5]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:58:10.910Z Evaluation of Visual Grounding performance on the test split of the ReferItGame dataset, specifically measured using Precision at 0.5 IoU (Pr@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 | Visual Grounding | | Dataset (including any stated split) | ReferitGame (test) | | Metric | Pr@0.5 | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 26 of 26 | | 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 - [Pr@0.5](): Higher is better; unit: SCALAR. ## Recorded results | Method | Pr@0.5 | Result date | Report paper | | --- | --- | --- | --- | | RefTR | 0.7142 | Not recorded | 68fa6e0c1e209131cea3dad7 | | TransVG | 0.7073 | Not recorded | 68fa6e0c1e209131cea3dad7 | | SeqTR | 0.6966 | Not recorded | 68fa6e0c1e209131cea3dad7 | | LBYL-Net | 0.6747 | Not recorded | 68fa69b91e209131cea23bac | | LBYL-Net | 0.6548 | Not recorded | 68fa69b91e209131cea23bac | | SQC-Large | 0.646 | Not recorded | 68fa69b91e209131cea23bac | | ReSCL | 0.646 | Not recorded | 68fa6e0c1e209131cea3dad7 | | SQC-Base | 0.6433 | Not recorded | 68fa69b91e209131cea23bac | | RCCF | 0.6379 | Not recorded | 68fa69b91e209131cea23bac | | RCCF | 0.6379 | Not recorded | 68fa6e0c1e209131cea3dad7 | | DDPN+ | 0.63 | Not recorded | 68fa69b91e209131cea23bac | | DDPN | 0.63 | Not recorded | 68fa6e0c1e209131cea3dad7 | | LBYL-Net w/o LFC | 0.6259 | Not recorded | 68fa69b91e209131cea23bac | | FAOA | 0.6067 | Not recorded | 68fa6e0c1e209131cea3dad7 | | YOLO-VG | 0.593 | Not recorded | 68fa69b91e209131cea23bac | | Baseline | 0.5928 | Not recorded | 68fa69b91e209131cea23bac | | YOLO-VG | 0.5876 | Not recorded | 68fa69b91e209131cea23bac | | ZSGNet | 0.5863 | Not recorded | 68fa69b91e209131cea23bac | | ZSGNet | 0.5863 | Not recorded | 68fa6e0c1e209131cea3dad7 | | CITE | 0.3507 | Not recorded | 68fa69b91e209131cea23bac | | Similarity Net | 0.3454 | Not recorded | 68fa69b91e209131cea23bac | | SimilarityNet | 0.3454 | Not recorded | 68fa6e0c1e209131cea3dad7 | | VC | 0.3113 | Not recorded | 68fa69b91e209131cea23bac | | MAttNet | 0.2904 | Not recorded | 68fa69b91e209131cea23bac | | MAttNet | 0.2904 | Not recorded | 68fa6e0c1e209131cea3dad7 | | CMN | 0.2833 | Not recorded | 68fa69b91e209131cea23bac | ### RefTR Pr@0.5: 0.7142. Result ID: 698ff6f13a6c314849fd209e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Stage":"One-stage","Visual Encoder":"RN101","Time (ms)":"40"}. This record does not assert independent reproduction. ### TransVG Pr@0.5: 0.7073. Result ID: 698ff6f13a6c314849fd209d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Stage":"One-stage","Visual Encoder":"RN101","Time (ms)":"62"}. This record does not assert independent reproduction. ### SeqTR Pr@0.5: 0.6966. Result ID: 698ff6f13a6c314849fd209f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Stage":"One-stage","Visual Encoder":"DN53","Time (ms)":"50"}. This record does not assert independent reproduction. ### LBYL-Net Pr@0.5: 0.6747. Result ID: 698ddd67ffc3901de0993a11. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Visual Encoder":"DarkNet-53","Language Encoder":"BERT","Processing Stage":"One-stage","Time":"30 ms"}. This record does not assert independent reproduction. ### LBYL-Net Pr@0.5: 0.6548. Result ID: 698ddd67ffc3901de0993a10. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Visual Encoder":"DarkNet-53","Language Encoder":"LSTM","Processing Stage":"One-stage","Time":"28 ms"}. This record does not assert independent reproduction. ### SQC-Large Pr@0.5: 0.646. Result ID: 698ddd67ffc3901de0993a0d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"DarkNet-53","Language Encoder":"BERT","Processing Stage":"One-stage","Time":"36 ms"}. This record does not assert independent reproduction. ### ReSCL Pr@0.5: 0.646. Result ID: 698ff6f13a6c314849fd209c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Stage":"One-stage","Visual Encoder":"DN53","Time (ms)":"36"}. This record does not assert independent reproduction. ### SQC-Base Pr@0.5: 0.6433. Result ID: 698ddd67ffc3901de0993a0c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"DarkNet-53","Language Encoder":"BERT","Processing Stage":"One-stage","Time":"26 ms"}. This record does not assert independent reproduction. ### RCCF Pr@0.5: 0.6379. Result ID: 698ddd67ffc3901de0993a09. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"DLA-34","Language Encoder":"LSTM","Processing Stage":"One-stage","Time":"25 ms"}. This record does not assert independent reproduction. ### RCCF Pr@0.5: 0.6379. Result ID: 698ff6f13a6c314849fd209b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Stage":"One-stage","Visual Encoder":"DLA34","Time (ms)":"25"}. This record does not assert independent reproduction. ### DDPN+ Pr@0.5: 0.63. Result ID: 698ddd67ffc3901de0993a07. [Reporting paper]() Recorded evaluation configuration: {"Visual Encoder":"ResNet-101","Language Encoder":"LSTM","Processing Stage":"Two-stage"}. This record does not assert independent reproduction. ### DDPN Pr@0.5: 0.63. Result ID: 698ff6f13a6c314849fd2098. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Stage":"Two-stage","Visual Encoder":"RN101"}. This record does not assert independent reproduction. ### LBYL-Net w/o LFC Pr@0.5: 0.6259. Result ID: 698ddd67ffc3901de0993a0f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Visual Encoder":"DarkNet-53","Language Encoder":"LSTM","Processing Stage":"One-stage","Time":"25 ms"}. This record does not assert independent reproduction. ### FAOA Pr@0.5: 0.6067. Result ID: 698ff6f13a6c314849fd2099. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Stage":"One-stage","Visual Encoder":"DN53","Time (ms)":"23"}. This record does not assert independent reproduction. ### YOLO-VG Pr@0.5: 0.593. Result ID: 698ddd67ffc3901de0993a0b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"DarkNet-53","Language Encoder":"BERT","Processing Stage":"One-stage","Time":"38 ms"}. This record does not assert independent reproduction. ### Baseline Pr@0.5: 0.5928. Result ID: 698ddd67ffc3901de0993a0e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"DarkNet-53","Language Encoder":"LSTM","Processing Stage":"One-stage","Time":"24 ms"}. This record does not assert independent reproduction. ### YOLO-VG Pr@0.5: 0.5876. Result ID: 698ddd67ffc3901de0993a0a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"DarkNet-53","Language Encoder":"LSTM","Processing Stage":"One-stage","Time":"21 ms"}. This record does not assert independent reproduction. ### ZSGNet Pr@0.5: 0.5863. Result ID: 698ddd67ffc3901de0993a08. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"ResNet-50","Language Encoder":"LSTM","Processing Stage":"One-stage","Time":"25 ms"}. This record does not assert independent reproduction. ### ZSGNet Pr@0.5: 0.5863. Result ID: 698ff6f13a6c314849fd209a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Stage":"One-stage","Visual Encoder":"RN50"}. This record does not assert independent reproduction. ### CITE Pr@0.5: 0.3507. Result ID: 698ddd67ffc3901de0993a05. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"ResNet-101","Processing Stage":"Two-stage","Time":"196 ms"}. This record does not assert independent reproduction. ### Similarity Net Pr@0.5: 0.3454. Result ID: 698ddd67ffc3901de0993a04. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"ResNet-101","Processing Stage":"Two-stage","Time":"184 ms"}. This record does not assert independent reproduction. ### SimilarityNet Pr@0.5: 0.3454. Result ID: 698ff6f13a6c314849fd2097. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Stage":"Two-stage","Visual Encoder":"RN101","Time (ms)":"184"}. This record does not assert independent reproduction. ### VC Pr@0.5: 0.3113. Result ID: 698ddd67ffc3901de0993a03. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"VGG-16","Language Encoder":"LSTM","Processing Stage":"Two-stage"}. This record does not assert independent reproduction. ### MAttNet Pr@0.5: 0.2904. Result ID: 698ddd67ffc3901de0993a06. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Visual Encoder":"ResNet-101","Language Encoder":"LSTM","Processing Stage":"Two-stage","Time":"320 ms"}. This record does not assert independent reproduction. ### MAttNet Pr@0.5: 0.2904. Result ID: 698ff6f13a6c314849fd2096. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Stage":"Two-stage","Visual Encoder":"RN101","Time (ms)":"320"}. This record does not assert independent reproduction. ### CMN Pr@0.5: 0.2833. Result ID: 698ddd67ffc3901de0993a02. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Visual Encoder":"VGG-16","Language Encoder":"LSTM","Processing Stage":"Two-stage"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.