# Phrase Grounding on Flickr30k entities (val) Source: [Phrase Grounding on Flickr30k entities (val)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:01:24.735Z Evaluation of Phrase Grounding performance on the validation split of the Flickr30k entities dataset, measured using R@K metrics. 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 | Phrase Grounding | | Dataset (including any stated split) | Flickr30k entities (val) | | Metric | Recall@1 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 35 of 35 | | 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 - [Recall@1](): Higher is better; unit: PERCENT. - [Recall@5](): Higher is better; unit: PERCENT. - [Recall@10](): Higher is better; unit: PERCENT. - [Accuracy@0.5](): Higher is better; unit: PERCENT. ## Recorded results | Method | Recall@1 | Result date | Report paper | | --- | --- | --- | --- | | FIBER-B | 87.1 | Not recorded | 68fa6e011e209131cea3d5c4 | | FIBER | 87.1 | Not recorded | 68fa700a1e209131cea48c7f | | GLIP-L | 86.7 | Not recorded | 68fa6cd91e209131cea36418 | | GLIP | 86.7 | Not recorded | 68fa6e011e209131cea3d5c4 | | GLIP | 86.7 | Not recorded | 68fa700a1e209131cea48c7f | | FIBER-B (w/o C.G. VLP) | 86.2 | Not recorded | 68fa6e011e209131cea3d5c4 | | GLIP-T | 85.7 | Not recorded | 68fa6cd91e209131cea36418 | | GLIP | 85.7 | Not recorded | 68fa6e011e209131cea3d5c4 | | GLIP-T | 84.8 | Not recorded | 68fa6cd91e209131cea36418 | | GLIP-T | 84 | Not recorded | 68fa6cd91e209131cea36418 | | MDETR-ENB5†\* | 83.6 | Not recorded | 68fa6af51e209131cea2b5a9 | | MDETR-ENB5 | 83.6 | Not recorded | 68fa6cd91e209131cea36418 | | MDETR | 83.6 | Not recorded | 68fa6e011e209131cea3d5c4 | | MDETR | 83.6 | Not recorded | 68fa700a1e209131cea48c7f | | MDETR-ENB3†\* | 82.9 | Not recorded | 68fa6af51e209131cea2b5a9 | | MDETR-R101†\* | 82.5 | Not recorded | 68fa6af51e209131cea2b5a9 | | MDETR-RN101 | 82.5 | Not recorded | 68fa6cd91e209131cea36418 | | MDETR-R101†\* | 82.3 | Not recorded | 68fa6af51e209131cea2b5a9 | | MDETR-R101 | 79 | Not recorded | 68fa6af51e209131cea2b5a9 | | MDETR-R101 | 78.9 | Not recorded | 68fa6af51e209131cea2b5a9 | | KOSMOS-2 | 77.8 | Not recorded | 68fa700a1e209131cea48c7f | | DDPN | 72.8 | Not recorded | 68fa6af51e209131cea2b5a9 | | VisualBert+ | 70.4 | Not recorded | 68fa6af51e209131cea2b5a9 | | Visual-BERT | 70.4 | Not recorded | 68fa6e011e209131cea3d5c4 | | VisualBert | 70.4 | Not recorded | 68fa700a1e209131cea48c7f | | VisualBert | 68.1 | Not recorded | 68fa6af51e209131cea2b5a9 | ### FIBER-B Recall@1: 87.1. Result ID: 698ff3233a6c314849fd01e5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Image Backbone":"Swin-B","#Pretrain Images (fine-grained)":"860k"}. This record does not assert independent reproduction. ### FIBER Recall@1: 87.1. Result ID: 699118013a6c31484905a35d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Zero-shot":false}. This record does not assert independent reproduction. ### GLIP-L Recall@1: 86.7. Result ID: 698f45353a6c314849f796b2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Data":"FourODs, GoldG, Cap24M"}. This record does not assert independent reproduction. ### GLIP Recall@1: 86.7. Result ID: 698ff3233a6c314849fd01e4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Image Backbone":"Swin-L","#Pretrain Images (fine-grained)":"27M"}. This record does not assert independent reproduction. ### GLIP Recall@1: 86.7. Result ID: 699118013a6c31484905a35c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Zero-shot":false}. This record does not assert independent reproduction. ### FIBER-B (w/o C.G. VLP) Recall@1: 86.2. Result ID: 698ff3233a6c314849fd01e6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Image Backbone":"Swin-B","#Pretrain Images (fine-grained)":"860k","Note":"without coarse-grained VL pretraining"}. This record does not assert independent reproduction. ### GLIP-T Recall@1: 85.7. Result ID: 698f45353a6c314849f796b1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Data":"O365, GoldG, Cap4M"}. This record does not assert independent reproduction. ### GLIP Recall@1: 85.7. Result ID: 698ff3233a6c314849fd01e3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Image Backbone":"Swin-B","#Pretrain Images (fine-grained)":"860k"}. This record does not assert independent reproduction. ### GLIP-T Recall@1: 84.8. Result ID: 698f45353a6c314849f796b0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Data":"O365, GoldG"}. This record does not assert independent reproduction. ### GLIP-T Recall@1: 84. Result ID: 698f45353a6c314849f796af. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Data":"GoldG"}. This record does not assert independent reproduction. ### MDETR-ENB5†\* Recall@1: 83.6. Result ID: 698e474effc3901de09c41c4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"ANY-BOX-PROTOCOL","Backbone":"EfficientNet-B5","Pre-training":"COCO, VG, Flickr 30k"}. This record does not assert independent reproduction. ### MDETR-ENB5 Recall@1: 83.6. Result ID: 698f45353a6c314849f796ae. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Data":"GoldG+"}. This record does not assert independent reproduction. ### MDETR Recall@1: 83.6. Result ID: 698ff3233a6c314849fd01e2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Image Backbone":"EN-B5","#Pretrain Images (fine-grained)":"200k"}. This record does not assert independent reproduction. ### MDETR Recall@1: 83.6. Result ID: 699118013a6c31484905a35b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Zero-shot":false}. This record does not assert independent reproduction. ### MDETR-ENB3†\* Recall@1: 82.9. Result ID: 698e474effc3901de09c41c3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"ANY-BOX-PROTOCOL","Backbone":"EfficientNet-B3","Pre-training":"COCO, VG, Flickr 30k"}. This record does not assert independent reproduction. ### MDETR-R101†\* Recall@1: 82.5. Result ID: 698e474effc3901de09c41c2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"ANY-BOX-PROTOCOL","Backbone":"ResNet-101","Pre-training":"COCO, VG, Flickr 30k"}. This record does not assert independent reproduction. ### MDETR-RN101 Recall@1: 82.5. Result ID: 698f45353a6c314849f796ad. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Data":"GoldG+"}. This record does not assert independent reproduction. ### MDETR-R101†\* Recall@1: 82.3. Result ID: 698e474effc3901de09c41c7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"MERGED-BOXES-PROTOCOL","Backbone":"ResNet-101","Pre-training":"COCO, VG, Flickr 30k"}. This record does not assert independent reproduction. ### MDETR-R101 Recall@1: 79. Result ID: 698e474effc3901de09c41c6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"MERGED-BOXES-PROTOCOL","Backbone":"ResNet-101","Pre-training":"None"}. This record does not assert independent reproduction. ### MDETR-R101 Recall@1: 78.9. Result ID: 698e474effc3901de09c41c1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation Protocol":"ANY-BOX-PROTOCOL","Backbone":"ResNet-101","Pre-training":"None"}. This record does not assert independent reproduction. ### KOSMOS-2 Recall@1: 77.8. Result ID: 699118013a6c31484905a35e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Zero-shot":true}. This record does not assert independent reproduction. ### DDPN Recall@1: 72.8. Result ID: 698e474effc3901de09c41c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Protocol":"MERGED-BOXES-PROTOCOL","Backbone":"ResNet-101"}. This record does not assert independent reproduction. ### VisualBert+ Recall@1: 70.4. Result ID: 698e474effc3901de09c41c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Protocol":"ANY-BOX-PROTOCOL","Backbone":"ResNet-101"}. This record does not assert independent reproduction. ### Visual-BERT Recall@1: 70.4. Result ID: 698ff3233a6c314849fd01e1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Image Backbone":"ResNet-101","#Pretrain Images (fine-grained)":"120k"}. This record does not assert independent reproduction. ### VisualBert Recall@1: 70.4. Result ID: 699118013a6c31484905a35a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Zero-shot":false}. This record does not assert independent reproduction. ### VisualBert Recall@1: 68.1. Result ID: 698e474effc3901de09c41bf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Protocol":"ANY-BOX-PROTOCOL","Backbone":"ResNet-101"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.