arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2105.14517cs.AI
收藏

GeoQA:面向多模态数值推理的几何问答基准

GeoQA: A Geometric Question Answering Benchmark Towards Multimodal Numerical Reasoning

Sun Yat-sen University(中山大学);Shenzhen Campus of Sun Yat-sen University(中山大学深圳校区);Dark Matter AI Inc.(暗物智能科技有限公司);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学)

Jiaqi Chen, Jianheng Tang, Jinghui Qin, Xiaodan Liang, Lingbo Liu, Eric P. Xing, Liang Lin

摘要:

自动数学问题解决作为一个长期存在的人工智能基准,最近引起了越来越多的关注。在本文中,我们专注于解决几何问题,这需要全面理解文本描述,可视化图表和定理知识。然而,现有的方法高度依赖于手工规则,并且仅在小规模数据集上进行评估。因此,我们提出了一个几何问答数据集GeoQA,包含4,998个几何问题与相应的注释程序,说明了给定问题的解决过程。与另一个公开的数据集GeoS相比,GeoQA是25倍大,其中的程序注释可以为未来的显式和可解释的数值推理研究提供一个实用的测试平台。此外,我们引入了一个神经几何求解器(NGS)来解决几何问题,全面解析多模态信息和生成可解释的程序。我们进一步在NGS上添加多个自监督辅助任务,以增强跨模态语义表示。GeoQA上的大量实验验证了我们提出的NGS和辅助任务的有效性。然而,结果仍然明显低于人类的表现,这为未来的研究留下了很大的空间。我们的基准测试和代码发布在https://github.com/chen-judge/GeoQA上。

英文摘要:

Automatic math problem solving has recently attracted increasing attention as a long-standing AI benchmark. In this paper, we focus on solving geometric problems, which requires a comprehensive understanding of textual descriptions, visual diagrams, and theorem knowledge. However, the existing methods were highly dependent on handcraft rules and were merely evaluated on small-scale datasets. Therefore, we propose a Geometric Question Answering dataset GeoQA, containing 4,998 geometric problems with corresponding annotated programs, which illustrate the solving process of the given problems. Compared with another publicly available dataset GeoS, GeoQA is 25 times larger, in which the program annotations can provide a practical testbed for future research on explicit and explainable numerical reasoning. Moreover, we introduce a Neural Geometric Solver (NGS) to address geometric problems by comprehensively parsing multimodal information and generating interpretable programs. We further add multiple self-supervised auxiliary tasks on NGS to enhance cross-modal semantic representation. Extensive experiments on GeoQA validate the effectiveness of our proposed NGS and auxiliary tasks. However, the results are still significantly lower than human performance, which leaves large room for future research. Our benchmark and code are released at https://github.com/chen-judge/GeoQA .

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑