Knowledge Editing
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
90.69BLEU
4
Jun 2, 2026
16.6Boost Delta (%)
4
May 8, 2026
59.253M-hop Success Rate
4
Apr 8, 2026
89.2Edit Success
3
Apr 21, 2026
25.7LG
3
Mar 4, 2026
84.7S Score
3
Mar 4, 2026
37.28Reliability
3
Feb 26, 2026
4,065Reliability
3
Feb 26, 2026
24.5Reliability
3
Feb 26, 2026
4,065Reliability
3
Feb 26, 2026
42.53Reliability
3
Feb 26, 2026
4,159Reliability
3
Feb 26, 2026
4,455Reliability
3
Feb 26, 2026
45.76Reliability
3
Feb 26, 2026
—
—Primary metric
0
May 8, 2026