MileBench: Benchmarking MLLMs in Long Context
This addresses a critical limitation for researchers and developers in evaluating MLLMs, though it is incremental as it focuses on benchmarking rather than model improvement.
The paper tackles the problem of unclear effectiveness of Multimodal Large Language Models (MLLMs) in real-world, long-context, and multi-image tasks by introducing MileBench, a benchmark that reveals most open-source MLLMs struggle in these scenarios, with performance gaps widening as image count increases.
Despite the advancements and impressive performance of Multimodal Large Language Models (MLLMs) on benchmarks, their effectiveness in real-world, long-context, and multi-image tasks is unclear due to the benchmarks' limited scope. Existing benchmarks often focus on single-image and short-text samples, and when assessing multi-image tasks, they either limit the image count or focus on specific task (e.g time-series captioning), potentially obscuring the performance challenges of MLLMs. To address these limitations, we introduce MileBench, a pioneering benchmark designed to test the MultImodal Long-contExt capabilities of MLLMs. This benchmark comprises not only multimodal long contexts, but also multiple tasks requiring both comprehension and generation. We establish two distinct evaluation sets, diagnostic and realistic, to systematically assess MLLMs' long-context adaptation capacity and their ability to complete tasks in long-context scenarios. Our experimental results, obtained from testing 22 models, revealed that while the closed-source GPT-4o outperforms others, most open-source MLLMs struggle in long-context situations. Interestingly, the performance gap tends to widen with an increase in the number of images. We strongly encourage an intensification of research efforts towards enhancing MLLMs' long-context capabilities, especially in scenarios involving multiple images.