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

Table Question Answering on WikiTQ

91.84Accuracy

GPT-5-nano

Updated 20d ago

Evaluation Results

Method	Links
GPT-5-nano 2026.01		91.84
GPT-5-nano 2026.01		89.8
GPT-oss-20b 2026.01		85.71
GPT-oss-20b 2026.01		85.71
Formula-R1-Plus 2025.05		82.54
GPT-4.1-nano 2026.01		81.63
TableMaster 2025.05		78.13
SS-CoT 2025.05		76.8
DeepSeek-R1 2025.11		75.76
TIDE DP&Agent 2025.05		75
TabAF 2025.05		74.72
Qwen2.5-Coder 2025.05		74.5
Table-R1 (w/ RE-SFT & TARPO) 2025.05		73.87
Norm-DP&Agent 2025.05		73.65
TableLLM (w/ SFT) 2025.05		73.02
TableGPT2 2025.05		71.45
GPT-4.1-nano 2026.01		71.43
QWQ 2025.05		70.5
VD-LB (HQ Correct vs. LQ Wrong) 2026.03		69.9
VD-LB (HQ Correct vs. LQ Wrong) 2026.03		69.6
Specialist SOTA 2025.11		69.1
RE-TableLLM (w/ RE-SFT) 2025.05		69.06
Qwen2.5-Instruct 2025.05		68.55
GPT-4 2025.11		68.4
GPT-4o 2025.05		68.4
VD-LF (HQ vs. LQ) 2026.03		68.2
VD-LB (HQ Correct vs. LQ Wrong) 2026.03		68.15
GPT-4.1 2025.11		68.14
ReAcTable 2025.05		68
Ministral-3-8B 2026.01		67.35
CHAIN-OF-TABLE 2024.01		67.31
Chain-of-Table 2025.05		67.31
VD-LF (HQ vs. LQ) 2026.03		67.15
Formula-R1 2025.05		67.05
Baseline 2026.03		66.4
TableLLM 2025.05		66.3
SFT (HQ Correct Only) 2026.03		66.05
SFT (HQ Correct Only) 2026.03		65.9
Dater 2025.05		65.9
VD-LF (HQ vs. LQ) 2026.03		65.4
Table-R1 (w/ RE-SFT & TARPO) 2025.05		65.27
Binder 2025.05		64.6
TableLLM (w/ SFT) 2025.05		64.16
RE-TableLLM (w/ RE-SFT) 2025.05		63.52
Table-R1 (w/ RE-SFT & TARPO) 2025.05		63.51
TableLLM (w/ SFT) 2025.05		63.37
Ministral-3-8B 2026.01		63.27
SFT (HQ Correct Only) 2026.03		62.85
OmniTab 2025.05		62.8
Table-R1 (w/ RE-SFT & TARPO) 2025.05		61.69
Dater 2024.01		61.48
TableGPT2 2025.05		61.42
CFMS 2026.04		61.42
TableGPT2 2025.05		61.42
RE-TableLLM (w/ RE-SFT) 2025.05		61.03
End-to-End QA 2024.01		60.59
Chain-of-Thought 2024.01		60.43
Few-Shot QA 2024.01		60.33
CHAIN-OF-TABLE 2024.01		59.94
Chain-of-Table 2026.04		59.94
CHAIN-OF-TABLE 2025.05		59.94
TAPEX-Large 2025.05		59.1
QWEN3-8B 2026.01		58.16
RE-TableLLM (w/ RE-SFT) 2025.05		57.55
Liu et al. (2021) 2022.01		57.5
QWEN3-8B 2026.01		57.14
TAMO 2025.11		56.93
Binder 2024.01		56.74
Binder 2026.04		56.74
Table-R1 (w/ RE-SFT & TARPO) 2025.05		56.51
QWEN3-8B 2026.01		55.61
TableLLM (w/ SFT) 2025.05		55.5
QWEN3-8B 2026.01		55.1
Binder 2024.01		54.88
Prompt tuning 2025.11		53.71
TableLLM 2025.05		53.59
Chain-of-Thought 2024.01		53.48
Chain-of-Thought 2026.04		53.48
GPT-3.5 2025.11		53.13
QWEN3-8B 2026.01		53.06
Text-to-SQL 2024.01		52.9
Text-to-SQL 2026.04		52.9
Dater 2024.01		52.81
Dater 2026.04		52.81
Few-Shot QA 2024.01		52.56
Few-Shot QA 2026.04		52.56
VD-LB (HQ Correct vs. LQ Wrong) 2026.03		52.45
Text-to-SQL 2024.01		52.42
VD-LF (HQ vs. LQ) 2026.03		52.15
End-to-End QA 2024.01		51.84
End-to-End QA 2026.04		51.84
VD-LF (HQ vs. LQ) 2026.03		51.7
RE-TableLLM (w/ RE-SFT) 2025.05		51.57
VD-LF (HQ vs. LQ) 2026.03		51.45
VD-LB (HQ Correct vs. LQ Wrong) 2026.03		51.35
Table-R1 (w/ RE-SFT & TARPO) 2025.05		51.35
VD-LB (HQ Correct vs. LQ Wrong) 2026.03		50.6
Table-R1 (w/ RE-SFT & TARPO) 2025.05		49.31
T5-3B 2022.01		49.29
QWEN3-8B 2026.01		48.98

Showing 100 of 149 rows