iTBLS: A Dataset of Interactive Conversations Over Tabular Information

TL;DR

This paper introduces iTBLS, a new dataset of interactive conversations over tables from academic papers, and proposes a question-answering reformulation approach that improves performance on tabular tasks.

Contribution

The paper presents a novel dataset for interactive table conversations and a reformulation framework that enhances task performance over existing methods.

Findings

Improved sequence-to-sequence model performance on iTBLS tasks.

Up to 13% accuracy improvement in text-to-table tasks.

Up to 16% BERTScore improvement over prior state-of-the-art.

Abstract

This paper introduces Interactive Tables (iTBLS), a dataset of interactive conversations that focuses on natural-language manipulation of tabular information sourced from academic pre-prints on ArXiv. The iTBLS dataset consists of three types of tabular tasks -- interpretation, modification, and generation. Interpretation focuses on tabular understanding, modification focuses on manipulating tabular information, and generation focuses on the addition of new natural-language evidence. In addition, the paper presents a novel framework that reformulates tabular operations as question-answering, where an appropriate question is formulated based on the nature of interaction and the question is answered using the user request as evidence. The developed approach results in an improvement on all tasks on a sequence-to-sequence modeling baseline on iTBLS. In addition, the question-answering-based reformulation is applied to datasets from prior work for the text-to-table task where textual paragraphs are summarized into tables. The novel approach results in up to 13% improvement in Exact-Match accuracy and up to 16% improvement in BERTScores compared to the prior state-of-the-art.