AutoBench-V: Can Large Vision-Language Models Benchmark Themselves?

TL;DR

AutoBench-V introduces an automated, flexible framework that uses LVLMs and text-to-image models to evaluate large vision-language models' capabilities without extensive human effort.

Contribution

The paper presents AutoBench-V, a novel automated benchmarking framework that enables on-demand, flexible evaluation of LVLMs using self-generated visual data.

Findings

Effective evaluation of nine LVLMs across multiple capabilities.

AutoBench-V reduces human effort in benchmarking.

Framework demonstrates reliability and flexibility.

Abstract

Large Vision-Language Models (LVLMs) have become essential for advancing the integration of visual and linguistic information. However, the evaluation of LVLMs presents significant challenges as the evaluation benchmark always demands lots of human cost for its construction, and remains static, lacking flexibility once constructed. Even though automatic evaluation has been explored in textual modality, the visual modality remains under-explored. As a result, in this work, we address a question: "Can LVLMs themselves be used to benchmark each other in the visual automatically domain?". We introduce AutoBench-V, an automated framework for serving evaluation on demand, i.e., benchmarking LVLMs based on specific aspects of model capability. AutoBench-V leverages text-to-image models to generate relevant image samples and then utilizes LVLMs to orchestrate visual question-answering (VQA) tasks, completing the evaluation process efficiently and flexibly. Through an extensive evaluation of nine popular LVLMs across five demanded user inputs (i.e., evaluation capabilities), the framework shows effectiveness and reliability.