TY - JOUR
AU - Ji, Rongrong
AB - Abstract:Multimodal Large Language Model (MLLM) relies on the powerful LLM to perform multimodal tasks, showing amazing emergent abilities in recent studies, such as writing poems based on an image. However, it is difficult for these case studies to fully reflect the performance of MLLM, lacking a comprehensive evaluation. In this paper, we fill in this blank, presenting the first comprehensive MLLM Evaluation benchmark MME. It measures both perception and cognition abilities on a total of 14 subtasks. In order to avoid data leakage that may arise from direct use of public datasets for evaluation, the annotations of instruction-answer pairs are all manually designed. The concise instruction design allows us to fairly compare MLLMs, instead of struggling in prompt engineering. Besides, with such an instruction, we can also easily carry out quantitative statistics. A total of 30 advanced MLLMs are comprehensively evaluated on our MME, which not only suggests that existing MLLMs still have a large room for improvement, but also reveals the potential directions for the subsequent model optimization.
TI - MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models
JF - Computing Research Repository
DO - 10.48550/arxiv.2306.13394
DA - 2023-06-23
UR - https://www.deepdyve.com/lp/arxiv-cornell-university/mme-a-comprehensive-evaluation-benchmark-for-multimodal-large-language-0i0KhU2mdd
VL - 2023
IS - 2306
DP - DeepDyve
ER -