Table-based Fact Verification
13 papers with code • 1 benchmarks • 2 datasets
Verifying facts given semi-structured data.
To this end, we construct a large-scale dataset called TabFact with 16k Wikipedia tables as the evidence for 118k human-annotated natural language statements, which are labeled as either ENTAILED or REFUTED.
To be able to use long examples as input of BERT models, we evaluate table pruning techniques as a pre-processing step to drastically improve the training and prediction efficiency at a moderate drop in accuracy.
TAPEX addresses the data scarcity challenge via guiding the language model to mimic a SQL executor on the diverse, large-scale and high-quality synthetic corpus.
AttesTable at SemEval-2021 Task 9: Extending Statement Verification with Tables for Unknown Class, and Semantic Evidence Finding
This paper describes our approach for Task 9 of SemEval 2021: Statement Verification and Evidence Finding with Tables.
From one perspective, our system conducts masked salient token prediction to enhance the model for alignment and reasoning between the table and the statement.
Logic-level Evidence Retrieval and Graph-based Verification Network for Table-based Fact Verification
Specifically, we first retrieve logic-level program-like evidence from the given table and statement as supplementary evidence for the table.
Fact verification based on structured data is challenging as it requires models to understand both natural language and symbolic operations performed over tables.
UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models
Structured knowledge grounding (SKG) leverages structured knowledge to complete user requests, such as semantic parsing over databases and question answering over knowledge bases.
The table-based fact verification task has recently gained widespread attention and yet remains to be a very challenging problem.
We propose Binder, a training-free neural-symbolic framework that maps the task input to a program, which (1) allows binding a unified API of language model (LM) functionalities to a programming language (e. g., SQL, Python) to extend its grammar coverage and thus tackle more diverse questions, (2) adopts an LM as both the program parser and the underlying model called by the API during execution, and (3) requires only a few in-context exemplar annotations.