</> Code with Papers by Syntology

Browse › Natural Language Processing › Machine Translation › WMT2014 English-French

Machine Translation archive 2025-07-28

WMT2014 English-French Benchmark (Machine Translation)

57 rows 54 with code listed 4 metrics Dataset page

Machine translation is the task of translating a sentence in a source language to a different target language.

Approaches for machine translation can range from rule-based to statistical to neural-based. More recently, encoder-decoder attention-based architectures like BERT have attained major improvements in machine translation.

One of the most popular datasets used to benchmark machine translation systems is the WMT family of datasets. Some of the most commonly used evaluation metrics for machine translation systems include BLEU, METEOR, NIST, and others.

The archive carries no text for this table; the description above is the archive's text for the task Machine Translation. archive 2025-07-28

Over time archive 2025-07-28

The chart needs JavaScript; the table below carries every value.

Direction inferred from the metric name, not from the archive: BLEU score (higher is better). Not inferred (points only, no best-so-far line): SacreBLEU, Hardware Burden, Operations per network pass. Points are placed at the row's paper date; 56 of 57 rows carry one.

Results archive 2025-07-28

Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.

Paper Code Ran Syntology Report
1 Transformer+BT (ADMIN init) 46.444.4 ✓ Paper Code 2020 8 of 9 ran · 1 unverified report
2 Noisy back-translation 45.643.8180G ✓ Paper Code 2018 1 of 1 ran · 0 unverified report
3 mRASP+Fine-Tune 44.341.7 ✓ Paper Code 2020 linked, not harvested report
4 Transformer + R-Drop 43.95 – Paper Code 2021 4 of 6 ran · 2 unverified report
5 Transformer (ADMIN init) 43.841.8 – Paper Code 2020 8 of 9 ran · 1 unverified report
6 Admin 43.8 – Paper Code 2020 4 of 5 ran · 1 unverified report
7 BERT-fused NMT 43.78 ✓ Paper Code 2020 linked, not harvested report
8 MUSE(Paralllel Multi-scale Attention) 43.5 – Paper Code 2019 0 of 1 ran · 1 unverified report
9 T5 43.4 ✓ Paper Code 2019 21 of 31 ran · 10 unverified report
10 Local Joint Self-attention 43.3 – Paper Code 2019 linked, not harvested report
11 Depth Growing 43.2724G – Paper Code 2019 linked, not harvested report
12 Transformer Big 43.255G – Paper Code 2018 linked, not harvested report
13 DynamicConv 43.2 – Paper Code 2019 linked, not harvested report
14 TaLK Convolutions 43.2 – Paper Code 2020 6 of 7 ran · 1 unverified report
15 LightConv 43.1 – Paper Code 2019 linked, not harvested report
16 FLOATER-large 42.7 – Paper Code 2020 3 of 6 ran · 3 unverified report
17 OmniNetP 42.6 – Paper Code 2021 linked, not harvested report
18 Transformer Big + MoS 42.1 – Paper Code 2018 linked, not harvested report
19 T2R + Pretrain 42.1 – Paper Code 2021 3 of 8 ran · 5 unverified report
20 Synthesizer (Random + Vanilla) 41.85 – Paper Code 2020 1 of 1 ran · 0 unverified report
21 Hardware Aware Transformer 41.8 – Paper Code 2020 linked, not harvested report
22 Transformer (big) + Relative Position Representations 41.5 – Paper Code 2018 13 of 23 ran · 10 unverified report
23 Stack 4-layer RNNSearch + Dual Learning + Deliberation Network 41.5 – Paper – 2017 no code linked report
24 Weighted Transformer (large) 41.4 – Paper Code 2017 2 of 2 ran · 0 unverified report
25 ConvS2S (ensemble) 41.3 – Paper Code 2017 linked, not harvested report
26 Evolved Transformer Big 41.3 – Paper Code 2019 linked, not harvested report
27 RNMT+ 41.0132G2.81G – Paper Code 2018 linked, not harvested report
28 Transformer Big 41.023G2300000000.0G ✓ Paper Code 2017 610 of 946 ran · 336 unverified report
29 Evolved Transformer Base 40.6 – Paper Code 2019 linked, not harvested report
30 ResMLP-12 40.6 – Paper Code 2021 5 of 7 ran · 2 unverified report
31 MoE 40.56142G – Paper Code 2017 4 of 6 ran · 2 unverified report
32 Transformer 40.5 – Paper Code 2019 linked, not harvested report
33 ConvS2S 40.46143G – Paper Code 2017 linked, not harvested report
34 ResMLP-6 40.3 – Paper Code 2021 5 of 7 ran · 2 unverified report
35 TransformerBase + AutoDropout 40 – Paper Code 2021 linked, not harvested report
36 GNMT+RL 39.9279G – Paper Code 2016 23 of 46 ran · 23 unverified report
37 Lite Transformer 39.6 – Paper Code 2020 linked, not harvested report
38 Deep-Att + PosUnk 39.2119G – Paper Code 2016 linked, not harvested report
39 Rfa-Gate-arccos 39.2 – Paper – 2021 no code linked report
40 Transformer Base 38.123G330000000.0G – Paper Code 2017 610 of 946 ran · 336 unverified report
41 LSTM6 + PosUnk 37.5 – Paper Code 2014 linked, not harvested report
42 PBMT 37 – – – not matched report
43 SMT+LSTM5 36.5 – Paper Code 2014 14 of 25 ran · 11 unverified report
44 RNN-search50* 36.2 – Paper Code 2014 21 of 44 ran · 23 unverified report
45 Deep-Att 35.9 – Paper Code 2016 linked, not harvested report
46 Deep Convolutional Encoder; single-layer decoder 35.7 – Paper Code 2016 linked, not harvested report
47 LSTM 34.8 – Paper Code 2014 14 of 25 ran · 11 unverified report
48 CSLM + RNN + WP 34.54 – Paper Code 2014 13 of 22 ran · 9 unverified report
49 FLAN 137B (zero-shot) 33.9 – Paper Code 2021 0 of 1 ran · 1 unverified report
50 FLAN 137B (few-shot, k=9) 33.8 – Paper Code 2021 0 of 1 ran · 1 unverified report
51 Regularized LSTM 29.03 – Paper Code 2014 2 of 6 ran · 4 unverified report
52 Unsupervised PBSMT 28.11 – Paper Code 2018 linked, not harvested report
53 PBSMT + NMT 27.6 – Paper Code 2018 linked, not harvested report
54 GRU+Attention 26.4 – Paper Code 2016 linked, not harvested report
55 SMT + iterative backtranslation (unsupervised) 26.22 – Paper Code 2018 linked, not harvested report
56 Unsupervised NMT + Transformer 25.14 – Paper Code 2018 linked, not harvested report
57 Unsupervised attentional encoder-decoder + BPE 14.36 – Paper Code 2017 6 of 6 ran · 0 unverified report

All 57 rows shown. 56 link to a paper page on this site; 6 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28

Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 27 rows have a graph line, from 22 distinct papers; 24 rows (20 papers) have at least one sample that ran. Counting each paper once: Syntology ran 764 of 1,209 samples; 445 unverified. Separately, 536 of those 1,209 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-25. Per-sample status is on the paper page.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections