LLaVA-UHD v2: an MLLM Integrating High-Resolution Semantic Pyramid via Hierarchical Window Transformer

Documenti analoghi

• LLaVA-UHD v3: Progressive Visual Compression for Efficient Native-Resolution Encoding in MLLMs
  di: Sun, Shichu, et al.
  Pubblicazione: (2025)
• LLaVA-UHD: an LMM Perceiving Any Aspect Ratio and High-Resolution Images
  di: Xu, Ruyi, et al.
  Pubblicazione: (2024)
• LLaVA-UHD v4: What Makes Efficient Visual Encoding in MLLMs?
  di: Fang, Kechen, et al.
  Pubblicazione: (2026)
• TG-LLaVA: Text Guided LLaVA via Learnable Latent Embeddings
  di: Yan, Dawei, et al.
  Pubblicazione: (2024)
• LLaVA-MoD: Making LLaVA Tiny via MoE Knowledge Distillation
  di: Shu, Fangxun, et al.
  Pubblicazione: (2024)