# Image Captioning on MS COCO 40,775 images (test) Source: [Image Captioning on MS COCO 40,775 images (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:38:21.767Z Evaluation of Image Captioning performance on the specific 40,775 image test subset of the MS COCO dataset, typically assessed using metrics like CIDEr, BLEU scores, ROUGE-L, and METEOR. 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 | Image Captioning | | Dataset (including any stated split) | MS COCO 40,775 images (test) | | Metric | CIDEr | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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 - [CIDEr](): Higher is better; unit: SCALAR. - [BLEU-4](): Higher is better; unit: SCALAR. - [BLEU-1](): Higher is better; unit: SCALAR. - [ROUGE-L](): Higher is better; unit: SCALAR. - [METEOR](): Higher is better; unit: SCALAR. - [BLEU-2](): Higher is better; unit: SCALAR. - [BLEU-3](): Higher is better; unit: SCALAR. ## Recorded results | Method | CIDEr | Result date | Report paper | | --- | --- | --- | --- | | X-Transformer | 133.5 | Not recorded | 68fa67501e209131cea1558a | | X-LAN | 132.8 | Not recorded | 68fa67501e209131cea1558a | | X-Transformer | 131.4 | Not recorded | 68fa67501e209131cea1558a | | X-LAN | 130.3 | Not recorded | 68fa67501e209131cea1558a | | HIP | 130.2 | Not recorded | 68fa67501e209131cea1558a | | AoANet | 129.6 | Not recorded | 68fa67501e209131cea1558a | | SGAE | 126.5 | Not recorded | 68fa67501e209131cea1558a | | GCN-LSTM | 126.5 | Not recorded | 68fa67501e209131cea1558a | | RFNet | 125.1 | Not recorded | 68fa67501e209131cea1558a | | Up-Down | 120.5 | Not recorded | 68fa67501e209131cea1558a | | LSTM-A | 118 | Not recorded | 68fa67501e209131cea1558a | | VGG+Score+DMSM+ft | 0.925 | Not recorded | 68fa624c1e209131ce9f8b1e | | VGG+Score+DMSM+ft | 0.912 | Not recorded | 68fa624c1e209131ce9f8b1e | | Human | 0.91 | Not recorded | 68fa624c1e209131ce9f8b1e | | Human | 0.854 | Not recorded | 68fa624c1e209131ce9f8b1e | ### X-Transformer CIDEr: 133.5. Result ID: 698d28928b517eb4331f6fdf. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"SENet-154"}. This record does not assert independent reproduction. ### X-LAN CIDEr: 132.8. Result ID: 698d28928b517eb4331f6fdd. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"SENet-154"}. This record does not assert independent reproduction. ### X-Transformer CIDEr: 131.4. Result ID: 698d28928b517eb4331f6fde. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-101"}. This record does not assert independent reproduction. ### X-LAN CIDEr: 130.3. Result ID: 698d28928b517eb4331f6fdc. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-101"}. This record does not assert independent reproduction. ### HIP CIDEr: 130.2. Result ID: 698d28928b517eb4331f6fdb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"SENet-154"}. This record does not assert independent reproduction. ### AoANet CIDEr: 129.6. Result ID: 698d28928b517eb4331f6fda. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-101"}. This record does not assert independent reproduction. ### SGAE CIDEr: 126.5. Result ID: 698d28928b517eb4331f6fd8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-101"}. This record does not assert independent reproduction. ### GCN-LSTM CIDEr: 126.5. Result ID: 698d28928b517eb4331f6fd9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-101"}. This record does not assert independent reproduction. ### RFNet CIDEr: 125.1. Result ID: 698d28928b517eb4331f6fd7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet+DenseNet+Inception"}. This record does not assert independent reproduction. ### Up-Down CIDEr: 120.5. Result ID: 698d28928b517eb4331f6fd6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"ResNet-101"}. This record does not assert independent reproduction. ### LSTM-A CIDEr: 118. Result ID: 698d28928b517eb4331f6fd5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"ResNet-152"}. This record does not assert independent reproduction. ### VGG+Score+DMSM+ft CIDEr: 0.925. Result ID: 698bd5fd7a967eddb771b302. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Number of reference captions":"40"}. This record does not assert independent reproduction. ### VGG+Score+DMSM+ft CIDEr: 0.912. Result ID: 698bd5fd7a967eddb771b300. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Number of reference captions":"5"}. This record does not assert independent reproduction. ### Human CIDEr: 0.91. Result ID: 698bd5fd7a967eddb771b303. [Reporting paper]() Recorded evaluation configuration: {"Number of reference captions":"40"}. This record does not assert independent reproduction. ### Human CIDEr: 0.854. Result ID: 698bd5fd7a967eddb771b301. [Reporting paper]() Recorded evaluation configuration: {"Number of reference captions":"5"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.