Beyond the Global Scores: Fine-Grained Token Grounding as a Robust Detector of LVLM Hallucinations

Documenti analoghi

• Overthinking Causes Hallucination: Tracing Confounder Propagation in Vision Language Models
  di: Shoby, Abin, et al.
  Pubblicazione: (2026)
• Localizing Before Answering: A Hallucination Evaluation Benchmark for Grounded Medical Multimodal LLMs
  di: Nguyen, Dung, et al.
  Pubblicazione: (2025)
• Interactive Medical Image Analysis with Concept-based Similarity Reasoning
  di: Huy, Ta Duc, et al.
  Pubblicazione: (2025)
• Enhanced Multimodal Video Retrieval System: Integrating Query Expansion and Cross-modal Temporal Event Retrieval
  di: Vo, Van-Thinh, et al.
  Pubblicazione: (2025)
• Any3DIS: Class-Agnostic 3D Instance Segmentation by 2D Mask Tracking
  di: Nguyen, Phuc, et al.
  Pubblicazione: (2024)