Seeing Time: 视觉-语言模型中的时间顺序推理与捷径偏差基准测试
Seeing Time: Benchmarking Chronological Reasoning and Shortcut Biases in Vision-Language Models
AI总结：
本文提出一个新基准，通过三个专门数据集评估视觉-语言模型在图像内和跨图像的时间顺序推理能力，并揭示模型常利用颜色等表面线索而非真正时间特征。
摘要：
近期视觉-语言模型（VLM）在解释复杂视觉语义方面取得了显著进展，但其时间顺序推理能力仍未得到充分探索。本文引入了一个新颖的基准，专门用于评估VLM如何感知和推理图像内及跨图像的时间顺序信息。与现有基于视频的基准（侧重于帧序列）不同，我们的工作深入探讨了时间判断的基本逻辑以及向多模态集成的扩展。为此，我们构建了三个专门数据集：一个包含跨越长时间历史周期的视觉相似物体，另一个按不同事件和物体类型分类，第三个将图像与时间敏感的新闻文本配对以实现跨模态对齐。通过大量实验，我们分析了模型是否在不同类别间表现出性能差异，并关键地探讨了它们是否依赖“错误捷径”（如图像颜色而非真正的时间特征）。我们的结果表明，尽管VLM显示出潜力，但它们经常利用灰度与彩色滤镜等表面线索来绕过真正的时间顺序推理。通过提供这些高质量数据集和严格的评估框架，我们提供了一个诊断工具，用于识别当前局限性并指导开发更稳健、逻辑更严密的多模态模型。源代码见 https://github.com/LuoRenqiang/ChronoVision。
英文摘要：
Recent advancements in Vision-Language Models (VLMs) have significantly enhanced their ability to interpret complex visual semantics, yet their capacity for chronological reasoning remains under-explored. In this paper, we introduce a novel benchmark specifically designed to evaluate how VLMs perceive and reason about chronological information within and across images. Unlike existing video-based benchmarks that focus on frame sequencing, our work delves into the underlying logic of chronological judgment and the expansion toward multimodal integration. To facilitate this, we construct three specialized datasets: one containing visually similar objects spanning long historical durations, another categorized by diverse event and object types, and a third pairing images with time-sensitive news text for cross-modal alignment. Through extensive experiments, we analyze whether models exhibit performance disparities across categories and, crucially, explore whether they rely on ``incorrect shortcuts'', such as image color rather than genuine chronological features. Our results reveal that while VLMs show promise, they frequently exploit superficial cues like grayscale versus color filters to bypass authentic chronological reasoning. By providing these high-quality datasets and a rigorous evaluation framework, we offer a diagnostic tool to identify current limitations and guide the development of more robust, logically grounded multimodal models. The source code is shown in https://github.com/LuoRenqiang/ChronoVision.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
