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

LLM Alignment on OpenAssistant

87.08GPT-4o Win Rate

DPPrefSyn

0.635223.077645.5267.9624May 29, 2026
Updated 1d ago

Evaluation Results

MethodLinks
2026.05
87.08
2026.05
86.52
2026.05
61.52
2026.05
60.25
2026.05
59.41
2026.05
59.13
2026.05
54.92
2026.05
52.11
2026.05
51.69
2026.05
48.17
2026.05
45.79
2026.05
44.1
2026.05
42.28
2026.05
37.08
2026.05
36.66
2026.05
36.24
2026.05
36.01
2026.05
33.85
2026.05
30.06
2026.05
16.15
2026.05
12.36
2026.05
12.05
2026.05
11.85
2026.05
11.04
2026.05
10.31
2026.05
9.86
2026.05
9.8
2026.05
9.63
2026.05
9.33
2026.05
9.1
2026.05
9.07
2026.05
8.96
2026.05
8.2
2026.05
7.3
2026.05
6.32
2026.05
6.18
2026.05
6.12
2026.05
6.04
2026.05
5.65
2026.05
4.75
2026.05
4.49
2026.05
4.46
2026.05
4.38
2026.05
4.29
2026.05
3.96