InfoTok: Information-Theoretic Regularization for Capacity-Constrained Shared Visual Tokenization in Unified MLLMs

Documenti analoghi

• InfoTok: Adaptive Discrete Video Tokenizer via Information-Theoretic Compression
  di: Ye, Haotian, et al.
  Pubblicazione: (2025)
• Visual Text Compression as Measure Transport
  di: Tang, Lv, et al.
  Pubblicazione: (2026)
• UniTok: A Unified Tokenizer for Visual Generation and Understanding
  di: Ma, Chuofan, et al.
  Pubblicazione: (2025)
• Mitigating Visual Hallucinations via Semantic Curriculum Preference Optimization in MLLMs
  di: Li, Yuanshuai, et al.
  Pubblicazione: (2025)
• UniWeTok: An Unified Binary Tokenizer with Codebook Size $\mathit{2^{128}}$ for Unified Multimodal Large Language Model
  di: Zhuang, Shaobin, et al.
  Pubblicazione: (2026)