Segmentation before Answering: Pixel Grounding for MLLM Visual Reasoning
For researchers working on multimodal large language models, this paper addresses the bottleneck of imprecise region-of-interest extraction by introducing segmentation masks, offering a more effective alternative to bounding boxes.
SegAnswer improves MLLM visual reasoning by using pixel-level segmentation masks instead of bounding boxes for region zoom-in, achieving consistent improvements across high-resolution perception, general perception, and hallucination benchmarks.
Recent advancements in Multimodal Large Language Models (MLLMs) have evolved from static perception to interleaved visual-language reasoning, often referred to as ``thinking with images''. A basic operation in this reasoning process is to zoom in on regions of interest (often represented with bounding boxes) to acquire finer visual details. In this paper, we propose \textbf{Seg}mentation before \textbf{Answer}ing (SegAnswer), which shifts the unit of zoom-in from the popular bounding box to pixel-level segmentation mask. By employing fine-grained masks to isolate the target area from cluttered environments, segmented visual input yields a more precise region of interest, effectively filtering out redundant background and interfering objects. Furthermore, the discrete patches of segmented visual input align more seamlessly with how MLLMs structure visual tokens via positional embeddings. In experiments, we evaluate SegAnswer across diverse benchmarks, including high-resolution perception, general perception, and hallucination. It achieves consistent improvements and also exhibits considerable performance on segmentation tasks, validating its capability for reliable pixel grounding.