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

Textual Understanding on AlpacaEval 2

54.7Length Constraint (LC)

Uni-DPO

3.7416.9730.243.43Jun 11, 2025
Updated 8d ago

Evaluation Results

MethodLinks
2025.06
54.752.7
2025.06
53.248
2025.06
52.540.5
2025.06
47.244.2
2025.06
46.842.6
2025.06
44.740.4
2025.06
43.741.7
2025.06
41.237.1
2025.06
28.528.4
2025.06
25.824.7
2025.06
23.820.5
2025.06
22.824.1
2025.06
20.918.4
2025.06
19.418.1
2025.06
15.513.3
2025.06
15.518.5
2025.06
5.73.7