NEMO: Can Multimodal LLMs Identify Attribute-Modified Objects?
This addresses a critical limitation in current multimodal LLMs for researchers and developers, though it is incremental as it builds on existing benchmarks and methods.
The paper tackles the problem of whether multimodal LLMs can identify objects modified by specific attributes, finding that they exhibit pronounced performance gaps and lag behind standalone vision encoders, with scaling up model size not consistently improving outcomes.
Multimodal Large Language Models (MLLMs) have made notable advances in visual understanding, yet their abilities to recognize objects modified by specific attributes remain an open question. To address this, we explore MLLMs' reasoning capabilities in object recognition, ranging from commonsense to beyond-commonsense scenarios. We introduce a novel benchmark, NEMO, which comprises 900 images of origiNal fruits and their corresponding attributE-MOdified ones; along with a set of 2,700 questions including open-, multiple-choice-, unsolvable types. We assess 26 recent open-sourced and commercial models using our benchmark. The findings highlight pronounced performance gaps in recognizing objects in NEMO and reveal distinct answer preferences across different models. Although stronger vision encoders improve performance, MLLMs still lag behind standalone vision encoders. Interestingly, scaling up the model size does not consistently yield better outcomes, as deeper analysis reveals that larger LLMs can weaken vision encoders during fine-tuning. These insights shed light on critical limitations in current MLLMs and suggest potential pathways toward developing more versatile and resilient multimodal models.