错觉-错觉：视觉语言模型在不存在错觉的地方看到错觉
The Illusion-Illusion: Vision Language Models See Illusions Where There Are None
AI总结：
研究通过给视觉语言模型呈现不应引发处理错误的‘错觉-错觉’，发现许多模型会误将其视为错觉，揭示了模型存在基本处理错误，此失败是文献中更广泛失败的一部分。
摘要：
错觉既有趣，又是认知科学、哲学和神经科学中有用的诊断工具。典型错觉展示了事物‘实际情况’与‘看起来的样子’之间的差距，有助于理解导致事物呈现方式的心理过程。错觉对研究人工系统也有用，许多研究探讨了感知计算模型是否会像人类一样陷入同样的错觉。本文通过呈现‘错觉-错觉’（常见错觉的类似物，不应引发处理错误）来研究当前视觉语言模型的基本处理错误。结果表明许多当前视觉语言系统会将这些‘错觉-错觉’误视为错觉，这种失败是文献中已讨论的更广泛失败的一部分。
英文摘要：
Illusions are entertaining, but they are also a useful diagnostic tool in cognitive science, philosophy, and neuroscience. A typical illusion shows a gap between how something `really is' and how something `appears to be', and this gap helps us understand the mental processing that led to how something appears to be. Illusions are also useful for investigating artificial systems, and much research has examined whether computational models of perception fall prey to the same illusions as people. Here, I invert the standard use of perceptual illusions to examine basic processing errors in current vision language models. I present these models with illusory-illusions, neighbors of common illusions that should not elicit processing errors. These include such things as perfectly reasonable ducks, crooked lines that truly are crooked, circles that seem to have different sizes because they are, in fact, of different sizes, and so on. I show that many current vision language systems mistakenly see these illusion-illusions as illusions. I suggest that such failures are part of broader failures already discussed in the literature.
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。
