# LEAF-QA: Locate, Encode & Attend for Figure Question Answering

## TL;DR

LEAF-QA introduces a large, complex dataset of real-world charts with questions, and proposes LEAF-Net, a novel architecture that advances multimodal chart question answering beyond previous datasets.

## Contribution

The paper presents LEAF-QA, a new large-scale dataset of real-world charts and questions, and introduces LEAF-Net, a novel architecture for multimodal chart question answering.

## Key findings

- LEAF-Net outperforms previous models on LEAF-QA, FigureQA, and DVQA.
- LEAF-QA is more complex and realistic than prior chart datasets.
- The architecture effectively localizes chart elements and encodes questions for improved QA performance.

## Abstract

We introduce LEAF-QA, a comprehensive dataset of $250,000$ densely annotated figures/charts, constructed from real-world open data sources, along with ~2 million question-answer (QA) pairs querying the structure and semantics of these charts. LEAF-QA highlights the problem of multimodal QA, which is notably different from conventional visual QA (VQA), and has recently gained interest in the community. Furthermore, LEAF-QA is significantly more complex than previous attempts at chart QA, viz. FigureQA and DVQA, which present only limited variations in chart data. LEAF-QA being constructed from real-world sources, requires a novel architecture to enable question answering. To this end, LEAF-Net, a deep architecture involving chart element localization, question and answer encoding in terms of chart elements, and an attention network is proposed. Different experiments are conducted to demonstrate the challenges of QA on LEAF-QA. The proposed architecture, LEAF-Net also considerably advances the current state-of-the-art on FigureQA and DVQA.