YARD: Y-Architecture Register Decoding for Efficient Hallucination Mitigation in Large Vision-Language Models 事件

YARD: Y-Architecture Register Decoding for Efficient Hallucination Mitigation in Large Vision-Language Models arXiv:2605.31429v1 Announce Type: new Abstract: Contrastive decoding (CD) seeks to mitigate hallucinations in Large Vision-Language Models (LVLMs) by contrasting the output distributions of a standard model and a visually degraded model. However, existing training-free CD methods suffer from sub-optimal degraded branches: completely dropping visual tokens is too extreme and induces lang