Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Object Hallucination Evaluation on HalluBench (test)
Loading...
76.6
Accuracy
Qwen3-VL-8B-Instruct + DUPL
73.272
74.136
75
75.864
Oct 1, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Qwen3-VL-8B-Instruct + DUPL
Model Size=8B, Optimiz...
2025.10
76.6
Qwen3-VL-4B-Instruct + DUPL
Model Size=4B, Optimiz...
2025.10
75.7
Qwen3-VL-8B-Instruct + GRPO
Model Size=8B, Optimiz...
2025.10
75.6
Qwen3-VL-4B-Instruct + GRPO
Model Size=4B, Optimiz...
2025.10
74.7
Qwen3-VL-8B-Instruct
Model Size=8B
2025.10
74.4
Qwen3-VL-4B-Instruct
Model Size=4B
2025.10
73.4
Feedback
Search any
task
Search any
task