测试视觉语言模型能否通过镜子识别自我，揭示其是否具备基于感知与行动的自我认知。
Mirror, Mirror on the Wall: Can VLM Agents Tell Who They Are at All?
- 设计3D镜像任务，让模型从反射中推断身体特征并正确选择目标
- 强模型能利用镜像信息指导行为，弱模型多误判或无法提取关键信息
- 揭示模型自指语言不等于真实自我认知，适合研究具身智能的可信性
在动物界，镜像自我识别是高级认知的典型指标，仅出现在部分物种中。我们探究类似功能是否在具身视觉-语言模型（VLM）代理中出现：它们能否在镜中识别自己？为此，我们引入一个受控的3D基准测试，要求第一人称VLM代理根据镜中反射推断隐藏的身体属性，并选择匹配目标，同时避免自我-他人混淆。为区分基于镜像的自我识别与捷径策略，我们测试了移除镜子、误导线索和遮挡反射的情况。还通过镜像探索、时间顺序、自我归因和推理-行动一致性评估决策过程。实验表明，只有较强模型表现出基于镜像的自我识别能力，能利用反射证据指导行动；而较弱模型虽观察镜子，却难以提取自我相关信息或产生误判。语言-视觉冲突进一步显示，仅靠自指语言不足以证明具身自我识别。总体而言，镜像评估可作为诊断工具，判断具身自我定位是否真正根植于感知与行动，而非先验知识、提示迎合或虚构。
原文摘要 · Abstract (English)
In the animal kingdom, mirror self-recognition is a canonical probe of higher-order cognition, emerging only in some species. We ask whether an analogous functional capability emerges in embodied vision-language model (VLM) agents: can they recognize themselves in a mirror? We introduce a controlled 3D benchmark where a first-person VLM agent must infer a hidden body attribute from its reflection and select the matching target, while avoiding self-other misattribution. To separate mirror-grounded self-identification from shortcuts, we test mirror removal, misleading cues, and occluded reflections. We also evaluate the decision process through mirror seeking, temporal ordering, self-attribution, and reasoning-action consistency. Our experiments show that mirror-based self-identification emerges mainly in stronger VLMs. These models can use reflected evidence for action, whereas weaker models often inspect the mirror but fail to extract self-relevant information or misattribute their reflection. Language-vision conflict further shows that self-referential language alone is not evidence of grounded self-identification. Overall, mirror-based evaluation provides a diagnostic for whether embodied self-grounding is causally rooted in perception and action rather than priors, prompt compliance, or confabulation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。