GENFIG1: Visual Summaries of Scholarly Work as a Challenge for Vision-Language Models

TL;DR

GENFIG1 is a new benchmark for evaluating AI models' ability to generate scientifically meaningful and visually coherent figures that summarize research ideas from scholarly papers.

Contribution

The paper introduces GENFIG1, a challenging benchmark for vision-language models to generate scientific figures that accurately and effectively communicate core research concepts.

Findings

Models struggle to generate accurate scientific figures.

The benchmark correlates well with expert human judgment.

Current models show significant room for improvement.

Abstract

In many science papers, "Figure 1" serves as the primary visual summary of the core research idea. These figures are visually simple yet conceptually rich, often requiring significant effort and iteration by human authors to get right, highlighting the difficulty of science visual communication. With this intuition, we introduce GENFIG1, a benchmark for generative AI models (e.g., Vision-Language Models). GENFIG1 evaluates models for their ability to produce figures that clearly express and motivate the central idea of a paper (title, abstract, introduction, and figure caption) as input. Solving GENFIG1 requires more than producing visually appealing graphics: the task entails reasoning for text-to-image generation that couples scientific understanding with visual synthesis. Specifically, models must (i) comprehend and grasp the technical concepts of the paper, (ii) identify the most salient ones, and (iii) design a coherent and aesthetically effective graphic that conveys those concepts visually and is faithful to the input. We curate the benchmark from papers published at top deep-learning conferences, apply stringent quality control, and introduce an automatic evaluation metric that correlates well with expert human judgments. We evaluate a suite of representative models on GENFIG1 and demonstrate that the task presents significant challenges, even for the best-performing systems. We hope this benchmark serves as a foundation for future progress in multimodal AI.