image

Text Template Tokens Are Implicit Semantic Registers in Diffusion Transformers

📄 Paper Trending (67⬆️): Text-to-image diffusion transformers (DiTs) jointly process text and image tokens, yet their internal computation during denoising remains poorly understood. We introduce a causal interpretability framework for modern large-scale DiTs that combines attention decomposition with targeted interventions...

Fonte: HuggingFace_Papers