Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Reasoning Faithfulness on DE driving data (held-out)
Loading...
61.4
Overall Consistency
Pinocchio
23.544
33.372
43.2
53.028
Jul 6, 2026
Overall Consistency
Consistency (E1)
Consistency (E2)
Consistency (E3)
Consistency (E4)
Consistency (E5)
Updated 19d ago
Evaluation Results
Method
Method
Links
Overall Consistency
Consistency (E1)
Consistency (E2)
Consistency (E3)
Consistency (E4)
Consistency (E5)
Pinocchio
Model=Ours
2026.07
61.4
69.2
86.4
97.3
82.5
71.1
VLM-Judge
Model=VLM-Judge
2026.07
57.5
76.9
89.9
97.4
78.3
64.8
ADE-Swap
Model=ADE-Swap
2026.07
43.9
55.6
75.5
95.6
73.4
58.3
ADE
Model=ADE
2026.07
43.4
51.7
72.7
95.5
73.6
59.1
SFT
Model=SFT
2026.07
27.7
48
68.5
95.1
60.1
43.2
ADE-Reason
Model=ADE-Reason
2026.07
25
45.7
67.5
95.2
51.6
38
Feedback
Search any
task
Search any
task
