Hidden in plain sight: VLMs overlook their visual representations

Documenti analoghi

• Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokens
  di: Qin, Yiming, et al.
  Pubblicazione: (2025)
• Hyperbolic Active Learning for Semantic Segmentation under Domain Shift
  di: Franco, Luca, et al.
  Pubblicazione: (2023)
• UnSAMv2: Self-Supervised Learning Enables Segment Anything at Any Granularity
  di: Yu, Junwei, et al.
  Pubblicazione: (2025)
• Reconstruction Alignment Improves Unified Multimodal Models
  di: Xie, Ji, et al.
  Pubblicazione: (2025)
• Compositional Chain-of-Thought Prompting for Large Multimodal Models
  di: Mitra, Chancharik, et al.
  Pubblicazione: (2023)