Your Large Vision-Language Model Only Needs A Few Attention Heads For Visual Grounding

Documenti analoghi

• See What You Are Told: Visual Attention Sink in Large Multimodal Models
  di: Kang, Seil, et al.
  Pubblicazione: (2025)
• Interpreting Attention Heads for Image-to-Text Information Flow in Large Vision-Language Models
  di: Kim, Jinyeong, et al.
  Pubblicazione: (2025)
• Real-Time Visual Attribution Streaming in Thinking Model
  di: Kang, Seil, et al.
  Pubblicazione: (2026)
• Interpreting vision transformers via residual replacement model
  di: Kim, Jinyeong, et al.
  Pubblicazione: (2025)
• FEAST: Fully Connected Expressive Attention for Spatial Transcriptomics
  di: Jeong, Taejin, et al.
  Pubblicazione: (2026)