List your company

Question Answering on WTQ (held-in)

80.34AccuracyGemini-2.5-Pro
-2.516818.994140.50562.0159Sep 21, 2025

Evaluation results

MethodLinks
Gemini-2.5-ProResolution=UNK2025.0980.34
OpenAI-o4-miniResolution=UNK2025.0978.2
Claude-3.5-SonnetResolution=UNK2025.0971.8
QVQ-72B-PreviewResolution=Dyn.2025.0968.2
Qwen2-VL-72BResolution=Dyn.2025.0965.7
GPT-4oResolution=UNK2025.0964.39
Qwen2-VL-7B-GRPOResolution=Dyn.2025.0953.2
Qwen2-VL-7B-Table-R1Resolution=Dyn.2025.0950.3
Qwen2-VL-7B-SFTResolution=Dyn.2025.0946.8
Qwen2-VL-7BResolution=Dyn.2025.0946.5
DeepSeek-VL2 4.5BResolution=Dyn.2025.0940.42
Qwen2-VL-2B-GRPOResolution=Dyn.2025.0935.3
Qwen2-VL-2B-Table-R1Resolution=Dyn.2025.0934.4
Qwen2-VL-2B-SFTResolution=Dyn.2025.0931
Qwen2-VL-2BResolution=Dyn.2025.0922.3
Table-LLaVA 13BResolution=3362025.0920.41
Qwen2-VL-7B-InsResolution=Dyn2025.0919.73
Monkey 7BResolution=8962025.0919.07
Table-LLaVA 7BResolution=3362025.0918.43
LLaVA-v1.5 7BResolution=3362025.091.24
mPLUG-Owl2 7BResolution=4482025.090.67