From Visuals to Vocabulary: Establishing Equivalence Between Image and Text Token Through Autoregressive Pre-training in MLLMs

Documenti analoghi

• VISTA: Enhancing Vision-Text Alignment in MLLMs via Cross-Modal Mutual Information Maximization
  di: Li, Mingxiao, et al.
  Pubblicazione: (2025)
• Advancing General Multimodal Capability of Vision-language Models with Pyramid-descent Visual Position Encoding
  di: Chen, Zhanpeng, et al.
  Pubblicazione: (2025)
• RLVMR: Reinforcement Learning with Verifiable Meta-Reasoning Rewards for Robust Long-Horizon Agents
  di: Zhang, Zijing, et al.
  Pubblicazione: (2025)
• Time Matters: Enhancing Pre-trained News Recommendation Models with Robust User Dwell Time Injection
  di: Jiang, Hao, et al.
  Pubblicazione: (2024)
• TS-LLaVA: Constructing Visual Tokens through Thumbnail-and-Sampling for Training-Free Video Large Language Models
  di: Qu, Tingyu, et al.
  Pubblicazione: (2024)