START: Spatial and Textual Learning for Chart Understanding

TL;DR

START introduces a novel approach combining spatial and textual learning to improve chart understanding in multimodal large language models, using a new dataset and benchmark for evaluation.

Contribution

The paper proposes START, a method integrating chart-element grounding and chart-to-code generation, along with a new dataset and benchmark for comprehensive chart understanding.

Findings

START achieves significant performance improvements over baseline models.

The START-Dataset enables effective training of spatial and textual chart understanding.

START surpasses previous state-of-the-art methods on benchmark evaluations.

Abstract

Chart understanding is crucial for deploying multimodal large language models (MLLMs) in real-world scenarios such as analyzing scientific papers and technical reports. Unlike natural images, charts pair a structured visual layout (spatial property) with an underlying data representation (textual property) -- grasping both is essential for precise, fine-grained chart reasoning. Motivated by this observation, we propose START, the Spatial and Textual learning for chART understanding. Specifically, we introduce (i) chart-element grounding and (ii) chart-to-code generation to strengthen an MLLM's understanding of both chart visual layout and data details. To facilitate spatial and textual learning, we propose the START-Dataset generated with a novel data-generation pipeline that first leverages an MLLM to translate real chart images into executable chart code, recovering the underlying data representation while preserving the visual distribution of real-world charts. We then evolve the code with a Large Language Model (LLM) to ascertain the positions of chart elements that capture the chart's visual structure, addressing challenges that existing methods cannot handle. To evaluate a model's ability to understand chart spatial structures, we propose the Chart Spatial understanding Benchmark (CS-Bench), filling a critical gap in comprehensive chart understanding evaluation. Leveraging spatial and textual learning, START delivers consistent gains across model sizes and benchmarks over the base models and surpasses prior state-of-the-art by a clear margin. Code, data and models will be publicly available.