GOBench: Benchmarking Geometric Optics Generation and Understanding of MLLMs

TL;DR

GOBench is a new benchmark that systematically evaluates Multi-modality Large Language Models' abilities in geometric optics generation and understanding, revealing significant challenges and gaps in current models.

Contribution

This paper introduces GOBench, the first comprehensive benchmark for assessing MLLMs' geometric optics capabilities through generation and understanding tasks.

Findings

Current models struggle with optical generation and understanding.

GPT-4o-Image shows limitations in generating authentic optical imagery.

Gemini-2.5Pro achieves only 37.35% accuracy in optical understanding.

Abstract

The rapid evolution of Multi-modality Large Language Models (MLLMs) is driving significant advancements in visual understanding and generation. Nevertheless, a comprehensive assessment of their capabilities, concerning the fine-grained physical principles especially in geometric optics, remains underexplored. To address this gap, we introduce GOBench, the first benchmark to systematically evaluate MLLMs' ability across two tasks: 1) Generating Optically Authentic Imagery and 2) Understanding Underlying Optical Phenomena. We curates high-quality prompts of geometric optical scenarios and use MLLMs to construct GOBench-Gen-1k dataset.We then organize subjective experiments to assess the generated imagery based on Optical Authenticity, Aesthetic Quality, and Instruction Fidelity, revealing MLLMs' generation flaws that violate optical principles. For the understanding task, we apply crafted evaluation instructions to test optical understanding ability of eleven prominent MLLMs. The experimental results demonstrate that current models face significant challenges in both optical generation and understanding. The top-performing generative model, GPT-4o-Image, cannot perfectly complete all generation tasks, and the best-performing MLLM model, Gemini-2.5Pro, attains a mere 37.35\% accuracy in optical understanding. Database and codes are publicly available at https://github.com/aiben-ch/GOBench.