Improve MLLM Benchmark Efficiency Through Interview
Farong Wen, Yijin Guo, Junying Wang, Jiahao Xiao, Yingjie Zhou, Ye Shen, Chunyi Li, Qi Jia, Zicheng Zhang · 2026
With the rapid development of Multimodal Large Language Models (MLLMs), their applications have expanded significantly, and numerous benchmark datasets have been proposed to evaluate their capabilities. However, full-coverage Q&A testing on large-scale data is resource-intensive and time-consuming. To address this issue, we propose the MLLM Interview (MITV) strategy, which aims to quickly obtain MLLM performance metrics by asking fewer questions. First, we constructed the interview dataset, which was built on an existing MLLM assessment dataset, by adding difficulty labels based on the performance of some typical MLLMs in this dataset. Second, we propose an MLLM Interview strategy, which obtains an initial performance situation of the large model by quizzing a small number of topics and then continuously tries to test the model’s limits. Through extensive experiments, the result shows that the MITV strategy proposed in this paper performs well on MLLM benchmark datasets, and it is able to obtain the model evaluation capability faster through a small number of questions and answers.