# Multimodal Machine Translation on Multi30k En-Fr (MSCOCO)

Source: [Multimodal Machine Translation on Multi30k En-Fr (MSCOCO)](<https://www.sota2.com/research/sota/multimodal-machine-translation-on-multi30k-en-fr-mscoco>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:49:49.248Z

Evaluation of Multimodal Machine Translation performance using the English-to-French translation task derived from the Multi30k dataset (which uses MSCOCO captions).

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 | Multimodal Machine Translation |
| Dataset (including any stated split) | Multi30k En-Fr (MSCOCO) |
| Metric | METEOR Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [METEOR Score](<https://www.sota2.com/research/sota/multimodal-machine-translation-on-multi30k-en-fr-mscoco.md?metric=METEOR%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | METEOR Score | Result date | Report paper |
| --- | --- | --- | --- |
| Gated Fusion | 70.51 | Not recorded | 68fa6a0d1e209131cea25ca5 |
| Imagination | 70.35 | Not recorded | 68fa6a0d1e209131cea25ca5 |
| Doubly-ATT | 70.25 | Not recorded | 68fa6a0d1e209131cea25ca5 |
| RMMT | 70.24 | Not recorded | 68fa6a0d1e209131cea25ca5 |
| Transformer-Tiny | 69.43 | Not recorded | 68fa6a0d1e209131cea25ca5 |
| Transformer-Small | 69.1 | Not recorded | 68fa6a0d1e209131cea25ca5 |
| Transformer-Base | 68.57 | Not recorded | 68fa6a0d1e209131cea25ca5 |
| DCCN | 65 | Not recorded | 68fa6a0d1e209131cea25ca5 |

### Gated Fusion

METEOR Score: 70.51. Result ID: 698e0602ffc3901de09a390c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25ca5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25ca5.md>)

[Implementation](<https://github.com/LividWo/Revisit-MMT>)

Recorded evaluation configuration: {"#Params":"2.8M","configuration":"Tiny"}. This record does not assert independent reproduction.

### Imagination

METEOR Score: 70.35. Result ID: 698e0602ffc3901de09a390b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25ca5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa637f1e209131ce9fee8b.md>)

Recorded evaluation configuration: {"#Params":"6.9M","configuration":"Tiny"}. This record does not assert independent reproduction.

### Doubly-ATT

METEOR Score: 70.25. Result ID: 698e0602ffc3901de09a390a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25ca5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63191e209131ce9fcb92.md>)

Recorded evaluation configuration: {"#Params":"3.2M","configuration":"Tiny"}. This record does not assert independent reproduction.

### RMMT

METEOR Score: 70.24. Result ID: 698e0602ffc3901de09a390d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25ca5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25ca5.md>)

[Implementation](<https://github.com/LividWo/Revisit-MMT>)

Recorded evaluation configuration: {"#Params":"2.9M","configuration":"Tiny"}. This record does not assert independent reproduction.

### Transformer-Tiny

METEOR Score: 69.43. Result ID: 698e0602ffc3901de09a3908.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25ca5.md>)

Recorded evaluation configuration: {"#Params":"2.6M"}. This record does not assert independent reproduction.

### Transformer-Small

METEOR Score: 69.1. Result ID: 698e0602ffc3901de09a3907.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25ca5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67051e209131cea13d76.md>)

[Implementation](<https://github.com/bert-nmt/bert-nmt>)

Recorded evaluation configuration: {"#Params":"36.4M"}. This record does not assert independent reproduction.

### Transformer-Base

METEOR Score: 68.57. Result ID: 698e0602ffc3901de09a3906.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25ca5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa703d1e209131cea49db1.md>)

[Implementation](<https://github.com/tensorflow/tensor2tensor>)

Recorded evaluation configuration: {"#Params":"49.0M"}. This record does not assert independent reproduction.

### DCCN

METEOR Score: 65. Result ID: 698e0602ffc3901de09a3909.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6a0d1e209131cea25ca5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68451e209131cea1aefd.md>)

[Implementation](<https://github.com/DeepLearnXMU/MM-DCCN>)

Recorded evaluation configuration: {"#Params":"16.9M"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.