VLM-SubtleBench: How Far Are VLMs from Human-Level Subtle Comparative Reasoning?

The ability to distinguish subtle differences between visually similar images is essential for diverse domains such as industrial anomaly detection, medical imaging, and aerial surveillance. While comparative reasoning benchmarks for visionlanguage models (VLMs) have recently emerged, they primarily focus on images with large, salient differences and fail to capture the nuanced reasoning required for real-world applications. In this work, we introduce VLM-SubtleBench 1 , a benchmark designed to evaluate VLMs on subtle comparative reasoning. Our benchmark covers ten difference types-Attribute, State, Emotion, Temporal, Spatial, Existence, Quantity, Quality, Viewpoint, and Action-and curate paired question-image sets reflecting these fine-grained variations. Unlike prior benchmarks restricted to natural image datasets, our benchmark spans diverse domains, including industrial, aerial, and medical imagery. Through extensive evaluation of both proprietary and open-source VLMs, we reveal systematic gaps between model and human performance across difference types and domains, and provide controlled analyses highlighting where VLMs' reasoning sharply deteriorates. Together, our benchmark and findings establish a foundation for advancing VLMs toward human-level comparative reasoning. Recently, vision-language models (VLMs) have shown remarkable progress toward artificial general intelligence (AGI), demonstrating promising results in various tasks, such as visual question answering (VQA) and scene description (Zhang et al., 2024 ). Yet, most progress has primarily centered on single visual inputs, e.g., an image or a video, while comparative tasks that require comparison over * Equal contribution. † Work done during an internship at KRAFTON.