TabSQLify: Enhancing Reasoning Capabilities of LLMs Through Table Decomposition
Open paper
Benchmark
Model
Rank
Results
table-based-fact-verification-on-tabfact
TabSQLify (col+row)
#11
Test: 79.5
semantic-parsing-on-wikitablequestions
TabSQLify (col+row)
#12
Accuracy (Test): 64.7
Rank counts only results with a code link.