DualToken: Towards Unifying Visual Understanding and Generation with Dual Visual Vocabularies

The differing representation spaces required for visual understanding and generation pose a challenge in unifying them within the autoregressive paradigm of large language models. A vision tokenizer trained for reconstruction excels at capturing low-level perceptual details, making it well-suited for visual generation but lacking high-level semantic representations for understanding tasks. Conversely, a vision encoder trained via contrastive learning aligns well with language but struggles to decode back into the pixel space for generation tasks. To bridge this gap, we propose DualToken, a method that unifies representations for both understanding and generation within a single tokenizer. However, directly integrating reconstruction and semantic objectives in a single tokenizer creates conflicts, leading to degraded performance in both reconstruction quality and semantic performance. Instead of forcing a single codebook to handle both semantic and perceptual information, DualToken disentangles them by introducing separate codebooks for high and low-level features, effectively transforming their inherent conflict into a synergistic relationship. As a result, DualToken achieves state-of-the-art performance in both reconstruction and semantic tasks while demonstrating remarkable effectiveness in downstream MLLM understanding and generation tasks. Notably, we also show that DualToken, as a unified tokenizer, surpasses the naive combination of two distinct types vision encoders, providing superior performance within a unified MLLM.

@article{song2025_2503.14324,
  title={ DualToken: Towards Unifying Visual Understanding and Generation with Dual Visual Vocabularies },
  author={ Wei Song and Yuran Wang and Zijia Song and Yadong Li and Haoze Sun and Weipeng Chen and Zenan Zhou and Jianhua Xu and Jiaqi Wang and Kaicheng Yu },
  journal={arXiv preprint arXiv:2503.14324},
  year={ 2025 }
}