Explanation Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
3.71PPL
7
Apr 21, 2026
4.5PPL
7
Apr 21, 2026
2.532BLEU-4
7
Mar 24, 2026
1.064BLEU-4
7
Mar 24, 2026
32.1ROUGE-1
7
Feb 26, 2026
23.4ROUGE-L
7
Feb 26, 2026
30.3ROUGE-1
7
Feb 26, 2026
2.98Grammar Score
7
Feb 26, 2026
2.99Grammar Score
7
Feb 26, 2026
4FMR
7
Feb 27, 2026
77FMR (Faithfulness Rate)
7
Feb 27, 2026
5FMR
7
Feb 27, 2026
4.936BLEU-4
6
Jun 30, 2026
91.7MoRF (Pixel Value) Insertion
6
May 19, 2026
61.4BLEU-4
6
May 6, 2026
1,788.6Nw Score
6
Feb 26, 2026
94.87BERTScore Precision
6
Feb 26, 2026
88.84BERTScore
6
Feb 26, 2026
87.67BERTScore
6
Feb 26, 2026
40.6BLEU-1
6
Feb 26, 2026
17.8NLI Faithfulness
5
May 29, 2026
21.7NLI Score
5
May 29, 2026
57.8NLI Score
5
May 29, 2026
0.37WER
5
Feb 26, 2026
100EE Score
4
Feb 26, 2026