Machine Translation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
2.2BLEU Score (Tokenized)
2
Feb 27, 2026
5.7BLEU (tokenized)
2
Feb 27, 2026
15.2BLEU Score
2
Feb 26, 2026
38.6BLEU
2
Feb 26, 2026
39.83BLEU
2
Feb 26, 2026
33.22BLEU
2
Feb 26, 2026
2.69Perplexity
2
Feb 26, 2026
2.6Perplexity
2
Feb 26, 2026
3.35Perplexity
2
Feb 26, 2026
17.6BLEU
2
Feb 26, 2026
25.5BLEU
2
Feb 26, 2026
1Human Rank
2
Feb 27, 2026
2Human Rank
2
Feb 27, 2026
0.25Reward
2
Feb 26, 2026
100Preference Score (NSL-MT)
1
May 7, 2026
100Preference Score (NSL-MT)
1
May 7, 2026
11.2Performance Loss
1
Mar 24, 2026
1BLEU Rank
1
Feb 27, 2026
0.2LE
1
Feb 26, 2026
0.8LE
1
Feb 26, 2026
0.7LE
1
Feb 26, 2026
43.8Win Rate
1
Feb 26, 2026
25.19spBLEU
1
Feb 26, 2026
30.9BLEU Score
1
Feb 26, 2026
36.5BLEU Score
1
Feb 26, 2026