Text-to-image diffusion models have demonstrated an unparalleled ability to generate high-quality, diverse images from a textual prompt. However, the internal representations learned by these models remain an enigma. In this work, we present Conceptor, a novel method to interpret the internal representation of a textual concept by a diffusion model. This interpretation is obtained by decomposing the concept into a small set of human-interpretable textual elements. Applied over the state-of-the-art Stable Diffusion model, Conceptor reveals non-trivial structures in the representations of concepts. For example, we find surprising visual connections between concepts, that transcend their textual semantics. We additionally discover concepts that rely on mixtures of exemplars, biases, renowned artistic styles, or a simultaneous fusion of multiple meanings of the concept. Through a large battery of experiments, we demonstrate Conceptor's ability to provide meaningful, robust, and faithful decompositions for a wide variety of abstract, concrete, and complex textual concepts, while allowing to naturally connect each decomposition element to its corresponding visual impact on the generated images. Our code will be available at: https://hila-chefer.github.io/Conceptor/

翻译：文本到图像扩散模型展现了从文本提示生成高质量、多样化图像的无与伦比的能力。然而，这些模型学习到的内部表征仍然是一个谜。在本研究中，我们提出了Conceptor，一种新颖的方法，用于解释扩散模型对文本概念的内部表征。这种解释通过将概念分解为一小组可被人类理解的文本元素来获得。应用于最先进的Stable Diffusion模型时，Conceptor揭示了概念表征中的非平凡结构。例如，我们发现概念之间存在超越其文本语义的惊人视觉联系。此外，我们还发现部分概念依赖于混合样例、偏见、知名艺术风格，或概念多重含义的同步融合。通过大量实验，我们证明了Conceptor能够为各种抽象、具体和复杂的文本概念提供有意义、稳健且忠实的分解，同时允许将每个分解元素自然地与其对生成图像的相应视觉影响联系起来。我们的代码将发布在：https://hila-chefer.github.io/Conceptor/