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

Source: [LEAF-QA: Locate, Encode & Attend for Figure Question Answering](<https://www.sota2.com/research/paper/68fa661e1e209131cea0e767>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T21:43:30.383Z

Authors: Ritwick Chaudhry, Sumit Shekhar, Utkarsh Gupta, Pranav Maneriker, Prann Bansal, Ajay Joshi

Published: 2019-07-30T12:39:06.000Z

## 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.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Chart Question Answering on FigurQA-D (test)](<https://www.sota2.com/research/sota/chart-question-answering-on-figurqa-d-test.md>): Chart Question Answering; FigurQA-D (test); Accuracy: 81.15.

- [Chart Question Answering on LeafQA-D (test)](<https://www.sota2.com/research/sota/chart-question-answering-on-leafqa-d-test.md>): Chart Question Answering; LeafQA-D (test); Accuracy: 67.42.

- [Chart Question Answering on DVQA-D (test)](<https://www.sota2.com/research/sota/chart-question-answering-on-dvqa-d-test.md>): Chart Question Answering; DVQA-D (test); Accuracy: 72.8.

- [Chart Question Answering on DVQA](<https://www.sota2.com/research/sota/chart-question-answering-on-dvqa.md>): Chart Question Answering; DVQA; Accuracy: 72.72.

- [Visual Question Answering on FigureQA Alternate color scheme (val)](<https://www.sota2.com/research/sota/visual-question-answering-on-figureqa-alternate-color-scheme-val.md>): Visual Question Answering; FigureQA Alternate color scheme (val); Accuracy: 81.15.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1907.12861>)

[PDF](<https://arxiv.org/pdf/1907.12861.pdf>)