D$^{3}$ToM: Decider-Guided Dynamic Token Merging for Accelerating Diffusion MLLMs

Documenti analoghi

• MuMA-ToM: Multi-modal Multi-Agent Theory of Mind
  di: Shi, Haojun, et al.
  Pubblicazione: (2024)
• Context Tokens are Anchors: Understanding the Repetition Curse in dMLLMs from an Information Flow Perspective
  di: Zhao, Qiyan, et al.
  Pubblicazione: (2026)
• Model Merging in LLMs, MLLMs, and Beyond: Methods, Theories, Applications and Opportunities
  di: Yang, Enneng, et al.
  Pubblicazione: (2024)
• ReGATE: Learning Faster and Better with Fewer Tokens in MLLMs
  di: Li, Chaoyu, et al.
  Pubblicazione: (2025)
• The Instinctive Bias: Spurious Images lead to Illusion in MLLMs
  di: Han, Tianyang, et al.
  Pubblicazione: (2024)