AI RESEARCH
WinTok: A Win-Win Hybrid Tokenizer via Decomposing Visual Understanding and Generation with Transferable Tokens
arXiv CS.CV
•
ArXi:2605.18115v1 Announce Type: new Building a unified visual tokenizer is essential for bridging the gap between visual understanding and generation. Yet existing approaches struggle with the inherent conflict between these tasks, as a single token space is forced to both high-level semantic abstraction and low-level pixel reconstruction. We propose WinTok, a concise hybrid tokenizer that achieves a win-win performance by explicitly decoupling the two objectives.