看见并不意味着定向：一个认知基础的基准揭示了多模态大语言模型在定向任务中的系统性失败
Seeing Isn't Orienting: A Cognitively Grounded Benchmark Reveals Systematic Orientation Failures in MLLMs Supplementary
AI总结：
本文提出DORI基准，通过分解定向为四个维度，揭示多模态模型在物体定向任务中的系统性不足，指出其依赖分类启发式而非几何推理。
摘要：
人类逐步学习物体定向，从识别物体方向到心理旋转再到物体间方向推理。现有视觉-语言基准将定向与位置和场景理解混为一谈。本文引入DORI，一个以认知为基础的分层基准，将物体定向作为主要目标。受人类定向认知阶段启发，DORI将定向分解为四个维度，每个维度在粗略（分类）和精细（度量）层面评估。DORI由13652张图像组成，涵盖14个来源，提供33656个多选问题，覆盖67个物体类别。其粗到细设计通过边界框隔离、标准化空间参考框架和结构化提示，隔离了物体识别难度、场景杂乱和语言模糊性等干扰因素。评估24个最先进的视觉-语言模型显示，表现良好的通用空间基准模型在物体导向任务中接近随机。最佳模型在粗略判断中达到54.2%，在精细判断中达到45.0%，最大的失败发生在复合旋转和物体间参考框架偏移。大粗到细差距揭示了模型依赖分类启发式而非几何推理，这一限制被现有基准所掩盖。这些结果表明，定向理解是多模态系统未解决的挑战，对机器人操作、3D场景重建和人机交互有影响。
英文摘要：
Humans learn object orientation progressively, from recognizing which way an object faces, to mentally rotating it, to reasoning about orientations between objects. Current vision-language benchmarks largely conflate orientation with position and general scene understanding. We introduce Discriminative Orientation Reasoning Intelligence (DORI), a cognitively grounded hierarchical benchmark that makes object orientation the primary target. Inspired by stages of human orientation cognition, DORI decomposes orientation into four dimensions, each evaluated at coarse (categorical) and granular (metric) levels. Composed from 13,652 images across 14 sources, DORI provides 33,656 multiple-choice questions covering 67 object categories in real-world and synthetic settings. Its coarse-to-granular design isolates orientation from confounds such as object recognition difficulty, scene clutter, and linguistic ambiguity via bounding-box isolation, standardized spatial reference frames, and structured prompts. Evaluating 24 state-of-the-art vision-language models shows a clear pattern: models that perform well on general spatial benchmarks are near-random on object-centric orientation tasks. The best models reach only 54.2% on coarse and 45.0% on granular judgments, with largest failures on compound rotations and shifts in inter-object reference frames. Large coarse-to-granular gaps reveal reliance on categorical heuristics rather than geometric reasoning, a limitation hidden by existing benchmarks. These results identify orientation understanding as an unsolved challenge for multimodal systems, with implications for robotic manipulation, 3D scene reconstruction, and human-AI interaction.
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。
