knowledge-infotabs.github.io - KNOWLEDGE_INFOTABS

home > domains > knowledge-infotabs.github.io

Example domain paragraphs

Reasoning about tabular information presents unique challenges to modern NLP approaches which largely rely on pre-trained contextualized embeddings of text. In this paper, we study these challenges through the problem of tabular natural language inference. We propose easy and effective modifications to how information is presented to a model for this task. We show via systematic experiments that these strategies substantially improve tabular inference performance.

TLDR; We propose effective modifications which are simple preprocessing of premsie table to enhance tabular reasoning by changing how tabular information is provided to a standard model. The Tabular Inference Problem Given a premise table, the task is to determine whether given hypothesis is true ( entailment ), false ( contradiction ), or undetermined ( neutral , i.e. tabular natural language inference. Below is an example from the INFOTABS dataset:

Here, H1 is entailed , H2 is contradiction and H3 is neutral

Links to knowledge-infotabs.github.io (4)

• vgupta123.github.io Vivek Gupta
• tabpert.github.io TabPert
• infotabs.github.io INFOTABS
• temptabqa.github.io TempTabQA