Look Carefully: Adaptive Visual Reinforcements in Multimodal Large Language Models for Hallucination Mitigation

Documenti analoghi

• Mitigating Hallucinations in Large Vision-Language Models without Performance Degradation
  di: Zhu, Xingyu, et al.
  Pubblicazione: (2026)
• Enhancing CLIP Robustness via Cross-Modality Alignment
  di: Zhu, Xingyu, et al.
  Pubblicazione: (2025)
• Principled Steering via Null-space Projection for Jailbreak Defense in Vision-Language Models
  di: Zhu, Xingyu, et al.
  Pubblicazione: (2026)
• Dynamic Multimodal Prototype Learning in Vision-Language Models
  di: Zhu, Xingyu, et al.
  Pubblicazione: (2025)
• Unsupervised Visual Chain-of-Thought Reasoning via Preference Optimization
  di: Zhao, Kesen, et al.
  Pubblicazione: (2025)