IKOD: Image-Attention-Guided Key-Value Merging against Object Hallucination in the Decoding of Large Vision-Language Models

arXiv:2508.03469 · cs.CV · Submitted 2025-08-05 · Read on arXiv

cs.CV

Submitted: 2025-08-05

Updated: 2026-11-09

Terminology

Sources

Related papers