TabFact

TabFact is a dataset that contains tables and corresponding natural language statements, used to evaluate the ability of models to perform reasoning and verification tasks involving tabular data.

Papers using TabFact (7)

TAPEX: Table Pre-training via Learning a Neural SQL Executor2021 · 90 citesBinding Language Models in Symbolic Languages2022 · 38 citesLarge Language Models are few(1)-shot Table Reasoners2022 · 7 citesProgram Enhanced Fact Verification with Verbalization and Graph Attention Network2020 · 1 citesTabSQLify: Enhancing Reasoning Capabilities of LLMs Through Table Decomposition2024NormTab: Improving Symbolic Reasoning in LLMs Through Tabular Data Normalization2024PoTable: Towards Systematic Thinking via Plan-then-Execute Stage Reasoning on Tables2024