Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General-domain Reasoning on HalluBench
Loading...
71
Accuracy
PAPO-7B
59.456
62.453
65.45
68.447
Oct 1, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
PAPO-7B
Base Model=PAPO-7B
2025.10
71
Qwen2.5-VL-7B + DUPL
Base Model=Qwen2.5-VL-...
2025.10
71
Qwen2.5-VL-7B + NoisyRollout
Base Model=Qwen2.5-VL-...
2025.10
70.1
VLAA-Thinker-7B
Base Model=VLAA-Thinke...
2025.10
70
Qwen2.5-VL-7B + GRPO
Base Model=Qwen2.5-VL-...
2025.10
68.6
R1-Onevision-7B
Base Model=R1-Onevisio...
2025.10
67.2
Qwen2.5-VL-3B + DUPL
Base Model=Qwen2.5-VL-...
2025.10
67
MM-Eureka-Qwen-7B
Base Model=MM-Eureka-Q...
2025.10
66.4
Qwen2.5-VL-3B + NoisyRollout
Base Model=Qwen2.5-VL-...
2025.10
66.1
Qwen2.5-VL-3B + GRPO
Base Model=Qwen2.5-VL-...
2025.10
65.5
Qwen2.5-VL-7B
Base Model=Qwen2.5-VL-7B
2025.10
65.2
OpenVLThinker-7B
Base Model=OpenVLThink...
2025.10
60
Qwen2.5-VL-3B
Base Model=Qwen2.5-VL-3B
2025.10
59.9
Feedback
Search any
task
Search any
task