Saved in:
Bibliographic Details
Main Authors: Ding, Xuanwen, Pan, Chengjun, Li, Zejun, Zhang, Jiwen, Wang, Siyuan, Wei, Zhongyu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.21389
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913861588746240
author Ding, Xuanwen
Pan, Chengjun
Li, Zejun
Zhang, Jiwen
Wang, Siyuan
Wei, Zhongyu
author_facet Ding, Xuanwen
Pan, Chengjun
Li, Zejun
Zhang, Jiwen
Wang, Siyuan
Wei, Zhongyu
contents Evaluating multimodal large language models (MLLMs) is increasingly expensive, as the growing size and cross-modality complexity of benchmarks demand significant scoring efforts. To tackle with this difficulty, we introduce AutoJudger, an agent-driven framework for efficient and adaptive benchmarking of MLLMs that tackles this escalating cost. AutoJudger employs the Item Response Theory (IRT) to estimate the question difficulty and an autonomous evaluation agent to dynamically select the most informative test questions based on the model's real-time performance. Specifically, AutoJudger incorporates two pivotal components: a semantic-aware retrieval mechanism to ensure that selected questions cover diverse and challenging scenarios across both vision and language modalities, and a dynamic memory that maintains contextual statistics of previously evaluated questions to guide coherent and globally informed question selection throughout the evaluation process. Extensive experiments on four representative multimodal benchmarks demonstrate that our adaptive framework dramatically reduces evaluation expenses, i.e. AutoJudger uses only 4% of the data to achieve over 90% ranking accuracy with the full benchmark evaluation on MMT-Bench.
format Preprint
id arxiv_https___arxiv_org_abs_2505_21389
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AutoJudger: An Agent-Driven Framework for Efficient Benchmarking of MLLMs
Ding, Xuanwen
Pan, Chengjun
Li, Zejun
Zhang, Jiwen
Wang, Siyuan
Wei, Zhongyu
Computation and Language
Evaluating multimodal large language models (MLLMs) is increasingly expensive, as the growing size and cross-modality complexity of benchmarks demand significant scoring efforts. To tackle with this difficulty, we introduce AutoJudger, an agent-driven framework for efficient and adaptive benchmarking of MLLMs that tackles this escalating cost. AutoJudger employs the Item Response Theory (IRT) to estimate the question difficulty and an autonomous evaluation agent to dynamically select the most informative test questions based on the model's real-time performance. Specifically, AutoJudger incorporates two pivotal components: a semantic-aware retrieval mechanism to ensure that selected questions cover diverse and challenging scenarios across both vision and language modalities, and a dynamic memory that maintains contextual statistics of previously evaluated questions to guide coherent and globally informed question selection throughout the evaluation process. Extensive experiments on four representative multimodal benchmarks demonstrate that our adaptive framework dramatically reduces evaluation expenses, i.e. AutoJudger uses only 4% of the data to achieve over 90% ranking accuracy with the full benchmark evaluation on MMT-Bench.
title AutoJudger: An Agent-Driven Framework for Efficient Benchmarking of MLLMs
topic Computation and Language
url https://arxiv.org/abs/2505.21389