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

Overall Performance on LoCoMo

48.7BLEU

Full Context

6.548817.491928.43539.3781Feb 3, 2026Feb 19, 2026Mar 8, 2026Mar 24, 2026Apr 10, 2026Apr 26, 2026May 13, 2026
Updated 20d ago

Evaluation Results

MethodLinks
2026.02
48.756.582.5
2026.02
46.655.386.3
2026.02
46.455.284.9
2026.02
45.653.479.4
2026.05
44.9151.35-
2026.05
40.6647.13-
2026.02
4047.673.4
2026.05
39.4245.77-
2026.02
38.945.563.3
2026.05
37.4844.15-
2026.02
34.641.165.3
2026.02
30.936.961.6
2026.05
29.5137.13-
2026.05
27.4936.07-
2026.05
25.5130.01-
2026.05
24.5433.05-
2026.05
23.6729.9-
2026.05
23.5931.45-
2026.05
20.7627.38-
2026.05
20.5527.83-
2026.05
15.2321.09-
2026.05
13.9119.63-
2026.05
9.4411.81-
2026.05
8.1710.1-