Explanation Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
1EE
4
Feb 26, 2026
100EE
4
Feb 26, 2026
1EE
4
Feb 26, 2026
1EE
4
Feb 26, 2026
1Explanation Effectiveness (EE)
4
Feb 26, 2026
98.7Politeness
4
Feb 26, 2026
97.4Politeness
4
Feb 26, 2026
97.3Politeness
4
Feb 26, 2026
78BS
3
Jun 16, 2026
74.1BS
3
Jun 16, 2026
0.94Linear Spike Score
3
Apr 3, 2026
13.14Mean Explanation Set Size
3
Feb 26, 2026
12.71Mean Explanation Set Size
3
Feb 26, 2026
3.48Mean Explanation Set Size
3
Feb 26, 2026
2.82Mean Explanation Set Size
3
Feb 26, 2026
0.118SFT (Mean Feature KL)
2
May 28, 2026
0.084SFT (Mean Feature KL)
2
May 28, 2026
24.5ROUGE-L
2
Apr 24, 2026
0.58BLEU-1
2
Mar 16, 2026
0.78Bias Score
2
Mar 4, 2026
0.677Diversity
2
Feb 26, 2026
14.473Time (s)
2
Feb 26, 2026
15.397Latency (s)
2
Feb 26, 2026
15.225Inference Time (s)
2
Feb 26, 2026
26.4Preference Rate (Ours)
1
Feb 26, 2026