Share your thoughts, 1 month free Claude Pro on usSee more

Fact Verification on TabFact (held-in)

85.02Accuracy

Gemini-2.5-Pro

Updated 22d ago

Evaluation Results

Method	Links
Gemini-2.5-Pro 2025.09		85.02
OpenAI-o4-mini 2025.09		84.7
QVQ-72B-Preview 2025.09		77.68
Qwen2-VL-72B 2025.09		73.45
Qwen2-VL-7B-Table-R1 2025.09		73.4
GPT-4o 2025.09		73.33
Qwen2-VL-7B-SFT 2025.09		73.2
Table-LLaVA 13B 2025.09		65
Qwen2-VL-2B-Table-R1 2025.09		60.9
Claude-3.5-Sonnet 2025.09		60.08
Table-LLaVA 7B 2025.09		59.85
Qwen2-VL-2B-SFT 2025.09		51.7
Qwen2-VL-7B-Ins 2025.09		40
Qwen2-VL-2B-GRPO 2025.09		22.7
Monkey 7B 2025.09		22.56
LLaVA-v1.5 7B 2025.09		18.9
DeepSeek-VL2 4.5B 2025.09		13.65
Qwen2-VL-7B-GRPO 2025.09		11.9
Qwen2-VL-2B 2025.09		8.9
mPLUG-Owl2 7B 2025.09		8.21
Qwen2-VL-7B 2025.09		7.4