Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Perception on TWI-oriented TreeBench (offline)
Loading...
71.1
Accuracy
RTWI
62.78
64.94
67.1
69.26
Feb 13, 2026
Accuracy
mIoU
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
mIoU
RTWI
Base Model=Qwen3-VL In...
2026.02
71.1
42
Deepconf
Base Model=Qwen3-VL In...
2026.02
69.8
38.9
Self-Cer.
Base Model=Qwen3-VL In...
2026.02
69.8
38.6
RTWI
Base Model=Qwen3-VL Th...
2026.02
68.5
36
SC
Base Model=Qwen3-VL In...
2026.02
68.4
38.6
CISC
Base Model=Qwen3-VL In...
2026.02
68.4
38.6
SC
Base Model=Qwen3-VL Th...
2026.02
66.4
32.5
CISC
Base Model=Qwen3-VL Th...
2026.02
66.4
32.5
Self-Cer.
Base Model=Qwen3-VL Th...
2026.02
65.8
32.5
Deepconf
Base Model=Qwen3-VL Th...
2026.02
64.4
32.9
Base
Base Model=Qwen3-VL In...
2026.02
64.4
-
Base
Base Model=Qwen3-VL Th...
2026.02
63.1
-
Feedback
Search any
task
Search any
task
