TextAtlas
No mentions found
This entity hasn't been tracked yet, or Iris is still building its knowledge base.
Related Articles from SNS
Residual Decoder Adapter: ID-Preserving Tokenizer Adaption for Autoregressive Text Rendering
arXiv:2606.01911v1 Announce Type: new Abstract: Visual Autoregressive (AR) models generate images by predicting discrete tokens that are decoded by a visual tokenizer. Despite demonstrating strong overall image generation ability, they still underperform on text rendering with blur strokes and disrupt letter shapes. In this work, we trace this limitation to the visual tokenizer, which struggles to reconstruct fine-grained detail.