When Token Pruning is Worse than Random: Understanding Visual Token Information in VLLMs

Documenti analoghi

• EntropyPrune: Matrix Entropy Guided Visual Token Pruning for Multimodal Large Language Models
  di: Wang, Yahong, et al.
  Pubblicazione: (2026)
• Self-Supervised Anatomical Consistency Learning for Vision-Grounded Medical Report Generation
  di: Yang, Longzhen, et al.
  Pubblicazione: (2025)
• AFiRe: Anatomy-Driven Self-Supervised Learning for Fine-Grained Representation in Radiographic Images
  di: Liu, Yihang, et al.
  Pubblicazione: (2025)
• M-IDoL: Information Decomposition for Modality-Specific and Diverse Representation Learning in Medical Foundation Model
  di: Liu, Yihang, et al.
  Pubblicazione: (2026)
• Visual Neural Decoding via Improved Visual-EEG Semantic Consistency
  di: Chen, Hongzhou, et al.
  Pubblicazione: (2024)