Definition Generation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
100Similarity Score
75
Mar 16, 2026
99.2Correctness
66
Mar 16, 2026
30.87BLEU
23
Feb 26, 2026
47.9BLEU
23
Feb 26, 2026
19.5BLEU
18
Feb 27, 2026
16.52BLEU
8
Feb 27, 2026
3Accuracy (a1)
3
Feb 26, 2026