AI Paper Atlas

Research map / Benchmarks / WMT14 En-Fr

WMT14 En-Fr leaderboard

28 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on WMT14 En-Fr. Best BLEU: 45.15 (F2-XEnDec, ICML 2021).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

WMT14 En-Fr · BLEU

28 papers, each with its best reported number for its own method; higher is better.

#BLEU ↑MethodPaperTopic
145.15F2-XEnDec
Transformer-big
Self-supervised and Supervised Joint Training for Resource-rich Machine Translation
ICML 2021 · Table 1
Self-supervised and contrastive
244.27PCformer (2-order)
Transformer-Big
Predictor-Corrector Enhanced Transformers with Exponential Moving Average Coefficient Learning
NeurIPS 2024 · Table 1
Attention and transformers
343.95HiCTL
XLM-R
On Learning Universal Representations Across Languages
ICLR 2021 · High-resource MT table
Recognition and long-tail
443.95Transformer + RD
Transformer big
R-Drop: Regularized Dropout for Neural Networks
NeurIPS 2021 · Table 2
Attention and transformers
543.9Hard fusion
Transformer
Understanding and Improving Encoder Layer Fusion in Sequence-to-Sequence Learning
ICLR 2021 · Table: Results of the proposed
Attention and transformers
643.9Transformer + MixtapeMixtape: Breaking the Softmax Bottleneck Efficiently
NeurIPS 2019 · Table 1
RNNs and normalization
743.9Mesh-TensorFlow Transformer, 2.89B
Transformer
Mesh-TensorFlow: Deep Learning for Supercomputers
NeurIPS 2018 · Table 2
RNNs and normalization
843.8THORTaming Sparsely Activated Transformer with Stochastic Experts
ICLR 2022 · Rich-resource results table
Mixture of experts
943.78BERT-fused model
Transformer
Incorporating BERT into Neural Machine Translation
ICLR 2020 · Table: BLEU scores of WMT'14 t
Recognition and long-tail
1042.5Transformer + Brier
Transformer
Language Generation with Strictly Proper Scoring Rules
ICML 2024 · Table: BLEU scores on WMT14 En
Language, safety and interpretability
18 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.