Machine Translation (EN→FR) on Multi30K 2016 (test)
63.7BLEUD2P-MMT (R)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| D2P-MMT (R)Method Category=Image-dependent Methods, Inference Image Source=Reconstructed, Ensemble=false2025.07 | 63.7 | — | |
| Noise-robustMethod Category=Image-dependent Methods, Ensemble=false2025.07 | 63.24 | — | |
| VALHALLA (M)Translation Setting=Multimodal, Ground-truth Visual Tokens usage=true, Model Averaging=true2022.05 | 63.1 | 81.8 | |
| VALHALLATranslation Setting=Text-Only, Ground-truth Visual Tokens usage=false, Model Averaging=true2022.05 | 63.1 | 81.8 | |
| VALHALLA*Method Category=Image-free Methods, Ensemble=true2025.07 | 63.1 | — | |
| VALHALLAMethod Category=Image-dependent Methods, Ensemble=false2025.07 | 63.1 | — | |
| D2P-MMT (A)Method Category=Image-dependent Methods, Inference Image Source=Authentic, Ensemble=false2025.07 | 63.04 | — | |
| Transformer-TinyMethod Category=Text-Only Transformer, Ensemble=false2025.07 | 62.84 | — | |
| IKD-MMTMethod Category=Image-free Methods, Ensemble=false2025.07 | 62.53 | — | |
| VALHALLA (M)Translation Setting=Multimodal, Ground-truth Visual Tokens usage=true, Model Averaging=false2022.05 | 62.4 | 81.4 | |
| VALHALLATranslation Setting=Text-Only, Ground-truth Visual Tokens usage=false, Model Averaging=false2022.05 | 62.3 | 81.4 | |
| Selective AttentionMethod Category=Image-dependent Methods, Ensemble=false2025.07 | 62.24 | — | |
| MMT-VQAMethod Category=Image-dependent Methods, Ensemble=false2025.07 | 62.24 | — | |
| RMMTTranslation Setting=Text-Only2022.05 | 62.1 | 81.3 | |
| RMMT*Method Category=Image-free Methods, Ensemble=true2025.07 | 62.1 | — | |
| Doubly-ATTMethod Category=Image-dependent Methods, Ensemble=false2025.07 | 61.99 | — | |
| Gated FusionTranslation Setting=Multimodal2022.05 | 61.7 | 81 | |
| Gated Fusion*Method Category=Image-dependent Methods, Ensemble=true2025.07 | 61.69 | — | |
| DCCNTranslation Setting=Multimodal2022.05 | 61.2 | 76.4 | |
| DCCNMethod Category=Image-dependent Methods, Ensemble=false2025.07 | 61.2 | — | |
| GMNMTTranslation Setting=Multimodal2022.05 | 60.9 | 74.9 | |
| CAP-ALLTranslation Setting=Multimodal2022.05 | 60.1 | 74.3 | |
| ImagiTTranslation Setting=Text-Only2022.05 | 59.7 | 74 | |
| ImagiTMethod Category=Image-free Methods, Ensemble=false2025.07 | 59.7 | — | |
| UVR-NMTTranslation Setting=Text-Only2022.05 | 58.3 | — | |
| UVR-MMTMethod Category=Image-free Methods, Ensemble=false2025.07 | 58.3 | — | |
| Gated FusionModality=Multimodal2022.12 | — | 73.1 | |
| Graph-MMTModality=Multimodal2022.12 | — | 74.1 | |
| mBART + MTModality=Text-only, Adapters=false2022.12 | — | 68.3 | |
| mBART + MTModality=Text-only, Adapters=true2022.12 | — | 79.9 | |
| TLM + MTModality=Text-only2022.12 | — | 76.3 | |
| Vanilla MTModality=Text-only2022.12 | — | 74.4 | |
| VGAMTModality=Multimodal2022.12 | — | 79.7 | |
| VTLM + MMTModality=Multimodal2022.12 | — | 75.9 |