Pretrained Vision Transformers (ViTs) such as DINOv2 and MAE provide generic image features that can be applied to a variety of downstream tasks such as retrieval, classification, and segmentation. However, such representations tend to focus on the most salient visual cues in the image, with no way to direct them toward less prominent concepts of interest. In contrast, Multimodal LLMs can be guided with textual prompts, but the resulting representations tend to be language-centric and lose their effectiveness for generic visual tasks. To address this, we introduce Steerable Visual Representations, a new class of visual representations, whose global and local features can be steered with natural language. While most vision-language models (e.g., CLIP) fuse text with visual features after encoding (late fusion), we inject text directly into the layers of the visual encoder (early fusion) via lightweight cross-attention. We introduce benchmarks for measuring representational steerability, and demonstrate that our steerable visual features can focus on any desired objects in an image while preserving the underlying representation quality. Our method also matches or outperforms dedicated approaches on anomaly detection and personalized object discrimination, exhibiting zero-shot generalization to out-of-distribution tasks.

翻译：预训练的视觉Transformer（ViT）模型，如DINOv2和MAE，提供了通用图像特征，可应用于检索、分类和分割等多种下游任务。然而，这类表征往往聚焦于图像中最显著的视觉线索，无法将其导向较不突出的感兴趣概念。相比之下，多模态大语言模型可通过文本提示进行引导，但由此产生的表征通常以语言为中心，会丧失其在通用视觉任务中的有效性。为此，我们提出可导向的视觉表征——一种新型视觉表征，其全局和局部特征均可通过自然语言进行导向。虽然大多数视觉语言模型（如CLIP）在编码后融合文本与视觉特征（晚期融合），但我们通过轻量级跨注意力机制将文本直接注入视觉编码器的各层（早期融合）。我们引入了衡量表征可导向性的基准，并证明我们的可导向视觉特征既能聚焦图像中任意目标物体，又能保持底层表征质量。此外，我们的方法在异常检测和个性化物体判别任务上达到或超越了专用方法的性能，并展现出对分布外任务的零样本泛化能力。