List your company

Video Text Visual Question Answering on RoadTextVQA (val)

78.97AccuracyVTAgent
22.955637.497852.0466.5822May 6, 2026

Evaluation results

MethodLinks
VTAgentYear=-2026.0578.9784.36
SFAYear=20252026.0561.1867.28
Qwen3-VL-8BYear=20252026.0553.3861.16
Qwen2.5-VL-7BYear=20252026.0550.3358.38
GAT-LargeYear=20252026.0550.2358.12
NVILAYear=20252026.0549.9857.22
TEA-LargeYear=20252026.0548.1454.85
Qwen2-VL-7BYear=20242026.0547.2355.34
GAT-BaseYear=20252026.0546.5453.78
TEA-BaseYear=20252026.0544.4351.69
Qwen3-VL-8B-ThinkingYear=20252026.0543.8653.09
InternVideo2.5-8BYear=20252026.0541.9649.05
Video-LLaVAYear=20242026.0530.8240.92
VideoLLaMA2Year=20242026.0525.1136.53