Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Text-to-SQL on Science Benchmark

59.53Execution Accuracy

KE-ICL

27.65435.929544.20552.4805May 7, 2025Jul 7, 2025Sep 7, 2025Nov 8, 2025Jan 9, 2026Mar 12, 2026May 13, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.05
59.53-
2025.05
59.5-
2025.05
59.2-
2025.05
58.2-
2025.05
57.9-
2026.05
56.5-
2025.05
56.2-
2025.05
56.2-
2025.05
55.5-
2026.05
54.84-
2025.05
54.6-
2025.05
53.8-
2025.05
53.8-
2026.05
53.62-
2026.05
53.17-
2025.05
52.8-
2025.05
52.5-
2026.01
51.861.9
2025.05
51.8-
2025.05
51.8-
2026.05
51.54-
2025.05
51.2-
2025.05
50.8-
2025.05
50.5-
2026.01
50.261.4
2026.05
50.2-
2026.05
49.16-
2026.05
48.32-
2026.05
46.88-
2026.05
46.48-
2026.01
45.253.4
2025.05
45.2-
2026.05
44.12-
2026.01
43.855.8
2026.05
43.14-
2025.05
43.1-
2026.05
41.13-
2026.05
39.79-
2026.05
39.29-
2026.05
38.22-
2025.05
36.8-
2026.05
36.12-
2026.05
34.62-
2026.05
34.44-
2026.05
33.11-
2026.05
31.89-
2026.05
29.76-
2026.05
28.88-