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

Navigation Question Answering on RAVEN-QA Real-World Robot Tasks

97.14Overall Accuracy

RAVEN (SigLIP)

1.064826.007450.9575.8926Jun 23, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
97.14
2026.06
93.33
2026.06
93.33
2026.06
92.86
2026.06
92.38
2026.06
90.47
2026.06
90
2026.06
90
2026.06
86.19
2026.06
85.71
2026.06
85.24
2026.06
84.76
2026.06
80.48
2026.06
76.67
2026.06
74.76
2026.06
67.62
2026.06
66.67
2026.06
66.67
2026.06
47.62
2026.06
45.71
2026.06
42.86
2026.06
35.24
2026.06
31.9
2026.06
30
2026.06
28.57
2026.06
14.76
2026.06
14.29
2026.06
4.76