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

Question Answering on TabMWP (Accuracy)

88.04Accuracy

zero-shot

Updated 25d ago

Evaluation Results

Method	Links
zero-shot 2026.06		88.04
GRAB 2026.06		87.78
GRAB 2026.06		87.01
Prompt tuning 2026.06		84.75
TAMO 2026.06		84.49
TableLLama 2026.06		80.05
GPT-5.4-mini 2026.06		72.06
NWCAD 2026.04		54.2
AdaCAD 2026.04		53.8
CoCoA 2026.04		53.2
CAD 2026.04		52.2
NWCAD 2026.04		50
AdaCAD 2026.04		49.12
CoCoA 2026.04		45.88
NWCAD 2026.04		36
zero-shot 2026.06		34.75
AdaCAD 2026.04		34.4
With-context 2026.04		31.8
CoCoA 2026.04		30.2
CAD 2026.04		30
With-context 2026.04		28.38
CAD 2026.04		27.2
With-context 2026.04		18.4
Baseline 2026.04		6.62
Baseline 2026.04		5.2
Baseline 2026.04		5