Prog-TAPAS: Enabling Table and Program Representation Consistency for Fact Verification

Siqi Wang, Chuang Zhang, Tingwen LiuIJCNN 2024 · OpenReview

Abstract

The table-based fact verification task is important yet challenging as it requires models to have both linguistic reasoning and symbolic reasoning ability. Existing studies mainly established two kinds of methods, program-enhanced methods and table-based pre-trained models, to excel at different reasoning aspects. In this paper, we propose a Program-enhanced Table Parser (Prog-TAPAS) to effectively combine these two methods. Leveraging a program enhanced module with program tables and synthesize program statements, we combine the strengths of symbolic reasoning into language-based inference models to support our fact verification model. Experiments show that our proposed framework outperforms all baseline systems with a considerable margin and achieves competitive results, on the TABFACT benchmark.
BibTeX
@inproceedings{wang2024prog,
  author = {Siqi Wang and Chuang Zhang and Tingwen Liu},
  title = {{Prog-TAPAS: Enabling Table and Program Representation Consistency for Fact Verification}},
  booktitle = {IJCNN},
  year = {2024},
  url = {https://openreview.net/forum?id=qG8heat7Wc},
}

Loading related content...

Figure