Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
AI Tutoring Dialogue Alignment on Socratic Mind
Loading...
70.47
Accuracy
MODPO
65.4052
66.7201
68.035
69.3499
Oct 1, 2025
Accuracy
Engagement
Updated 1d ago
Evaluation Results
Method
Method
Links
Accuracy
Engagement
MODPO
Backbone=Qwen2.5-7B-In...
2025.10
70.47
36
Single-Head DPO
Backbone=Qwen2.5-7B-In...
2025.10
70.4
44.6
MAH-DPO Acc Head
Backbone=Qwen2.5-7B-In...
2025.10
70.07
44.47
MAH-DPO Eng Head
Backbone=Qwen2.5-7B-In...
2025.10
69.53
44.8
MAH-DPO Ensemble
Backbone=Qwen2.5-7B-In...
2025.10
68.93
45.13
SFT
Backbone=Qwen2.5-7B-In...
2025.10
67.93
34.73
Base
Backbone=Qwen2.5-7B-In...
2025.10
65.6
32.2
Feedback
Search any
task
Search any
task