Spotlight and Shadow: Attention-Guided Dual-Anchor Introspective Decoding for MLLM Hallucination Mitigation

Documenti analoghi

• ASCD: Attention-Steerable Contrastive Decoding for Reducing Hallucination in MLLM
  di: Wang, Yujun, et al.
  Pubblicazione: (2025)
• Energy-Guided Decoding for Object Hallucination Mitigation
  di: Liu, Xixi, et al.
  Pubblicazione: (2025)
• Decoding by Perturbation: Mitigating MLLM Hallucinations via Dynamic Textual Perturbation
  di: Jia, Sihang, et al.
  Pubblicazione: (2026)
• MLLM can see? Dynamic Correction Decoding for Hallucination Mitigation
  di: Wang, Chenxi, et al.
  Pubblicazione: (2024)
• Visual Attention Drifts,but Anchors Hold:Mitigating Hallucination in Multimodal Large Language Models via Cross-Layer Visual Anchors
  di: Yang, Chengxu, et al.
  Pubblicazione: (2026)