List your company

Text-rich Visual Question Answering on TextVQA (Accuracy)

80.07AccuracyVanilla
50.003657.809365.61573.4207Jun 30, 2026

Evaluation results

MethodLinks
VanillaBackbone=Qwen3-VL-8B, TFLOPs=100%2026.0680.07
Op-SkipBackbone=Qwen3-VL-8B, TFLOPs=66%, N=202026.0678.45
APETBackbone=Qwen3-VL-8B, TFLOPs=66%2026.0678.36
VanillaBackbone=Qwen2.5-VL-7B, TFLOPs=100%2026.0677.77
V2DropBackbone=Qwen3-VL-8B, TFLOPs=66%2026.0676.44
ShortVBackbone=Qwen3-VL-8B, TFLOPs=58%, N=202026.0676.08
VSkipBackbone=Qwen3-VL-8B, TFLOPs=76%, N=202026.0675.98
V2DropBackbone=Qwen2.5-VL-7B, TFLOPs=55%2026.0672.26
Op-SkipBackbone=Qwen2.5-VL-7B, TFLOPs=55%, N=202026.0672.02
ShortVBackbone=Qwen2.5-VL-7B, TFLOPs=50%, N=202026.0671.93
APETBackbone=Qwen2.5-VL-7B, TFLOPs=55%2026.0670.9
VTWBackbone=Qwen2.5-VL-7B, TFLOPs=50%, K=162026.0665.43
VSkipBackbone=Qwen2.5-VL-7B, TFLOPs=67%, N=162026.0663.58
VanillaBackbone=LLaVA-1.5-7B, TFLOPs=100%2026.0658.21
VSkipBackbone=LLaVA-1.5-7B, TFLOPs=76%, N=202026.0657.21
VSkip+Backbone=LLaVA-1.5-7B, TFLOPs=76%, N=202026.0657.21
Op-SkipBackbone=LLaVA-1.5-7B, TFLOPs=66%, N=202026.0656.4
APETBackbone=LLaVA-1.5-7B, TFLOPs=66%2026.0656.36
V2DropBackbone=LLaVA-1.5-7B, TFLOPs=66%2026.0655.92
ShortVBackbone=LLaVA-1.5-7B, TFLOPs=55%, N=202026.0653.63
VTWBackbone=LLaVA-1.5-7B, TFLOPs=55%, K=162026.0652.17
VTWBackbone=Qwen3-VL-8B, TFLOPs=58%, K=162026.0651.16