Machine Translation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
5.496Mean Side-by-Side Score
3
Feb 26, 2026
5.504Side-by-Side Score (Mean)
3
Feb 26, 2026
0.2Reward
3
Feb 26, 2026
92# Improved
2
May 20, 2026
23.9Tie Rate (%)
2
May 18, 2026
16Tie Rate (%)
2
May 18, 2026
48.9Tie (%)
2
May 18, 2026
24Tie %
2
May 18, 2026
60.06Score
2
May 15, 2026
1.969PPL
2
May 7, 2026
0.157BLEU
2
Apr 29, 2026
91.05Mean DA
2
Apr 28, 2026
85.65Mean DA
2
Apr 28, 2026
82.55GRF
2
Apr 23, 2026
87.73Sent Score
2
Mar 27, 2026
54F-score
2
Mar 27, 2026
92F-score
2
Mar 27, 2026
100F-score
2
Mar 27, 2026
36.441BLEU (linewise)
2
Mar 18, 2026
39.66BLEU
2
Mar 4, 2026
28.48BLEU
2
Mar 4, 2026
20.8spBLEU (ar)
2
Feb 27, 2026
19.1spBLEU (ar)
2
Feb 27, 2026
8.9SacreBLEU (detokenized)
2
Feb 27, 2026
5.44BLEU
2
Feb 27, 2026