Seeing the Forest and the Trees: Solving Visual Graph and Tree Based Data Structure Problems using Large Multimodal Models

TL;DR

This paper evaluates large multimodal models' ability to solve graph and tree data structure problems from images, introducing a new benchmark dataset and analyzing model performance and limitations.

Contribution

It presents a novel benchmark dataset for visual graph and tree problems and assesses the performance of various large multimodal models on this dataset.

Findings

GPT-4o achieved 87.6% accuracy on tree problems.

Gemini 1.5 Flash achieved 56.2% accuracy on graph problems.

Model performance is significantly affected by structural and visual variations.

Abstract

Recent advancements in generative AI systems have raised concerns about academic integrity among educators. Beyond excelling at solving programming problems and text-based multiple-choice questions, recent research has also found that large multimodal models (LMMs) can solve Parsons problems based only on an image. However, such problems are still inherently text-based and rely on the capabilities of the models to convert the images of code blocks to their corresponding text. In this paper, we further investigate the capabilities of LMMs to solve graph and tree data structure problems based only on images. To achieve this, we computationally construct and evaluate a novel benchmark dataset comprising 9,072 samples of diverse graph and tree data structure tasks to assess the performance of the GPT-4o, GPT-4v, Gemini 1.5 Pro, Gemini 1.5 Flash, Gemini 1.0 Pro Vision, and Claude 3 model families. GPT-4o and Gemini 1.5 Flash performed best on trees and graphs respectively. GPT-4o achieved 87.6% accuracy on tree samples, while Gemini 1.5 Flash, achieved 56.2% accuracy on graph samples. Our findings highlight the influence of structural and visual variations on model performance. This research not only introduces an LMM benchmark to facilitate replication and further exploration but also underscores the potential of LMMs in solving complex computing problems, with important implications for pedagogy and assessment practices.