List your company
ResearchFormerly WizWand

Table Understanding on TableBench, WikiTQ, WikiSQL Aggregated

55.33Overall ScoreTable-R1 (w/ RE-SFT & TARPO)
18.753228.249137.74547.2409May 18, 2025
Updated 5mo ago

Evaluation results

MethodLinks
Table-R1 (w/ RE-SFT & TARPO)Base Model=Qwen3, Size=8B2025.0555.33
GPT-4o2025.0549.17
RE-TableLLM (w/ RE-SFT)Base Model=Qwen3, Size=8B2025.0548.98
Table-R1 (w/ RE-SFT & TARPO)Base Model=TeleChat2, Size=7B2025.0548.9
Table-R1 (w/ RE-SFT & TARPO)Base Model=Llama3.1, Size=8B2025.0547.97
Table-R1 (w/ RE-SFT & TARPO)Base Model=Qwen2, Size=7B2025.0546.42
TableGPT2Base Model=Qwen2.5, Size=72B2025.0546.2
Table-R1 (w/ RE-SFT & TARPO)Base Model=Qwen2.5, Size=3B2025.0545.74
QWQBase Model=Qwen2.5, Size=32B2025.0545.69
RE-TableLLM (w/ RE-SFT)Base Model=TeleChat2, Size=7B2025.0544.84
RE-TableLLM (w/ RE-SFT)Base Model=Llama3.1, Size=8B2025.0543.94
TableLLM (w/ SFT)Base Model=Qwen3, Size=8B2025.0543.2
RE-TableLLM (w/ RE-SFT)Base Model=Qwen2, Size=7B2025.0542.95
TableGPT2Base Model=Qwen2.5, Size=7B2025.0542.2
Table-R1 (w/ RE-SFT & TARPO)Base Model=DS-Coder, Size=7B2025.0541.29
RE-TableLLM (w/ RE-SFT)Base Model=Qwen2.5, Size=3B2025.0539.22
TableLLM (w/ SFT)Base Model=TeleChat2, Size=7B2025.0538.88
Table-R1 (w/ RE-SFT & TARPO)Base Model=CodeQwen, Size=7B2025.0538.58
Qwen2.5-CoderBase Model=Qwen2.5, Size=7B2025.0538.07
RE-TableLLM (w/ RE-SFT)Base Model=CodeQwen, Size=7B2025.0536.48
Qwen2.5-InstructBase Model=Qwen2.5, Size=7B2025.0536.14
RE-TableLLM (w/ RE-SFT)Base Model=DS-Coder, Size=7B2025.0536.13
TableLLM (w/ SFT)Base Model=Qwen2.5, Size=3B2025.0536.12
CHAIN-OF-TABLEBase Model=GPT3.52025.0534.63
TableLLM (w/ SFT)Base Model=Qwen2, Size=7B2025.0533.56
DS-liteBase Model=DS-Coder, Size=16B2025.0530
TableLLM (w/ SFT)Base Model=Llama3.1, Size=8B2025.0530
TableLLM (w/ SFT)Base Model=DS-Coder, Size=7B2025.0525.99
TableLLM (w/ SFT)Base Model=CodeQwen, Size=7B2025.0525.65
Yi-CoderBase Model=Yi, Size=9B2025.0522.21
TableLLMBase Model=CodeLlama, Size=13B2025.0520.16