Connect, Collapse, Corrupt: Learning Cross-Modal Tasks with Uni-Modal Data

Documenti analoghi

• Closing the Modality Gap for Mixed Modality Search
  di: Li, Binxu, et al.
  Pubblicazione: (2025)
• Just Shift It: Test-Time Prototype Shifting for Zero-Shot Generalization with Vision-Language Models
  di: Sui, Elaine, et al.
  Pubblicazione: (2024)
• Data or Language Supervision: What Makes CLIP Better than DINO?
  di: Liu, Yiming, et al.
  Pubblicazione: (2025)
• NegVQA: Can Vision Language Models Understand Negation?
  di: Zhang, Yuhui, et al.
  Pubblicazione: (2025)
• Why are Visually-Grounded Language Models Bad at Image Classification?
  di: Zhang, Yuhui, et al.
  Pubblicazione: (2024)