Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Physical Reasoning Visual Question Answering on CRAFT VQA
Loading...
39.4
Descriptive Accuracy
+EG+VAORA
26.088
29.544
33
36.456
Jul 7, 2026
Descriptive Accuracy
Counterfactual Accuracy
Enable Accuracy
Causal Accuracy
Prevent Accuracy
Overall Accuracy
Updated 17d ago
Evaluation Results
Method
Method
Links
Descriptive Accuracy
Counterfactual Accuracy
Enable Accuracy
Causal Accuracy
Prevent Accuracy
Overall Accuracy
+EG+VAORA
base model=Qwen3VL-8B-...
2026.07
39.4
48.2
46.3
45.9
50.5
46.06
+EG
base model=Qwen3VL-8B-...
2026.07
38.5
44.5
46
42.3
51.5
44.56
SFT
base model=Qwen3VL-8B
2026.07
37.7
46.9
43.9
44.7
47.5
44.14
Qwen3-VL-8B-Instruct
2026.07
26.6
41.7
40.7
42
40.7
38.34
Feedback
Search any
task
Search any
task