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

TableBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Table Question AnsweringTableBench 1.0 (test)
Accuracy48.46
136
Fact CheckingTableBench (test)
Accuracy85.42
136
Complex Tabular ReasoningTableBench
TB-DA68.8
45
Table Question AnsweringTableBench
EM62
40
Table ReasoningTABLEBENCH
Accuracy63.48
39
Table ReasoningTableBench
DP44.54
31
Table ReasoningTableBench
Pass@172.52
21
Numerical ReasoningTableBench (test)
Accuracy64.48
13
Table Chain of Thought ReasoningTableBench
Rge54.28
13
Symbolic Chain of Thought ReasoningTableBench
Rge1.99
13
Program of Thought ReasoningTableBench
Rge Score51.96
13
Data ProcessingTableBench
Rge52.18
13
Data AnalysisTableBench (test)
Accuracy26.63
10
Table Question AnsweringTableBench (test)
Overall Accuracy0.4887
10
STRINGTableBench
LLM Judge Accuracy32
9
Agentic Task SolvingTableBench
Pass@343
9
String ExtractionTableBench
Exact Match31.6
9
Table-QATableBench
Quality Score33.6
4
Showing 18 of 18 rows