DOCCI

DOCCI (Descriptions of Connected and Contrasting Images) is a collection of images paired with detailed descriptions. The descriptions explain the key elements of the images, as well as secondary information such as background, lighting, and settings. The images are specifically taken to help assess the precise visual

Papers using DOCCI (3)

TEVI: Text-Conditioned Editing of Visual Representations via Sparse Autoencoders for Improved Vision-Language Alignment2026VisionPangu: A Compact and Fine-Grained Multimodal Assistant with 1.7B Parameters2026The Telephone Game: Evaluating Semantic Drift in Unified Models2025