List your company
ResearchFormerly WizWand

Table Mathematical Reasoning on TabMWP (test)

99.57AccuracyTableMind++
40.113255.549170.98586.4209Mar 8, 2026Mar 23, 2026Apr 8, 2026Apr 24, 2026May 9, 2026May 25, 2026Jun 10, 2026
Updated 4mo ago

Evaluation results

MethodLinks
TableMind++Model Category=Tuning-based, Pass@1=true2026.0399.57
TableMindModel Category=Tuning-based, Pass@1=true2026.0399.27
Deepseek-R1Model Category=Open-source, Pass@1=true2026.0398.03
Qwen2.5-72B-InstructModel Category=Open-source, Pass@1=true2026.0397.45
Chain-of-TableModel Category=Training-free, Pass@1=true2026.0396.17
GPT-5Model Category=Proprietary, Pass@1=true2026.0396.12
Table-R1Model Category=Tuning-based, Pass@1=true2026.0396.02
MOCA-AGENTBackbone=Qwen3.6-27B, Method Category=Tool-augmented / agent-based methods2026.0696
Gemini-2.5-flashModel Category=Proprietary, Pass@1=true2026.0395.89
PoTableModel Category=Training-free, Pass@1=true2026.0395.76
CREATORBackbone=ChatGPT, Method Category=Tool-augmented / agent-based methods2026.0694.7
ChameleonBackbone=ChatGPT, Method Category=Tool-augmented / agent-based methods2026.0693.28
TaCoBackbone=TAPEX-large, Method Category=Fine-tuning approaches2026.0692.91
PoT+DocBackbone=ChatGPT, Method Category=Tool-augmented / agent-based methods2026.0692.69
Tab-CoTModel Category=Training-free, Pass@1=true2026.0392.21
Gemini-2.0-flashModel Category=Proprietary, Pass@1=true2026.0391.56
CoTBackbone=GPT-4, Method Category=Prompting / chain-of-thought with LLMs2026.0690.81
GPT-4.1Model Category=Proprietary, Pass@1=true2026.0390.34
CoS-PlanningBackbone=ChatGPT, Method Category=Tool-augmented / agent-based methods2026.0690
PoTBackbone=ChatGPT, Method Category=Prompting / chain-of-thought with LLMs2026.0689.49
CRITICBackbone=ChatGPT, Method Category=Tool-augmented / agent-based methods2026.0689
RetICLBackbone=Codex, Method Category=Prompting / chain-of-thought with LLMs2026.0688.51
CRAFTBackbone=ChatGPT, Method Category=Tool-augmented / agent-based methods2026.0688.4
CRITICBackbone=GPT-3, Method Category=Tool-augmented / agent-based methods2026.0687.6
TaCoBackbone=TAPEX-base, Method Category=Fine-tuning approaches2026.0686.12
TableLlama-7BModel Category=Tuning-based, Pass@1=true2026.0383.88
CoTBackbone=ChatGPT, Method Category=Prompting / chain-of-thought with LLMs2026.0682.03
PoT-SCBackbone=Codex, Method Category=Prompting / chain-of-thought with LLMs2026.0681.8
TableGPT2-7BModel Category=Tuning-based, Pass@1=true2026.0380.43
CRITICBackbone=Llama-2-70B, Method Category=Tool-augmented / agent-based methods2026.0675
ToRABackbone=Llama-2-70B, Method Category=Fine-tuning approaches2026.0674
ToRA-CodeBackbone=CodeLlama-34B, Method Category=Fine-tuning approaches2026.0670.5
Qwen3-8BModel Category=Open-source, Pass@1=true2026.0360.86
ToRABackbone=Llama-2-13B, Method Category=Fine-tuning approaches2026.0647.2
ToRABackbone=Llama-2-7B, Method Category=Fine-tuning approaches2026.0642.4