When are Lemons Purple? The Concept Association Bias of Vision-Language Models

Documenti analoghi

• If CLIP Could Talk: Understanding Vision-Language Model Representations Through Their Preferred Concept Descriptions
  di: Esfandiarpoor, Reza, et al.
  Pubblicazione: (2024)
• Language-Pretraining-Induced Bias: A Strong Foundation for General Vision Tasks
  di: Luo, Yaxin, et al.
  Pubblicazione: (2026)
• Unified Vision-Language Modeling via Concept Space Alignment
  di: Qiu, Yifu, et al.
  Pubblicazione: (2026)
• Dreaming Out Loud: A Self-Synthesis Approach For Training Vision-Language Models With Developmentally Plausible Data
  di: AlKhamissi, Badr, et al.
  Pubblicazione: (2024)
• S-GRPO: Unified Post-Training for Large Vision-Language Models
  di: Yan, Yuming, et al.
  Pubblicazione: (2026)