{"ID":2854424,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.14304","arxiv_id":"2510.14304","title":"Watermarking for Factuality: Guiding Vision-Language Models Toward Truth via Tri-layer Contrastive Decoding","abstract":"Large Vision-Language Models (LVLMs) have recently shown promising results on various multimodal tasks, even achieving human-comparable performance in certain cases. Nevertheless, LVLMs remain prone to hallucinations -- they often rely heavily on a single modality or memorize training data without properly grounding their outputs. To address this, we propose a training-free, tri-layer contrastive decoding with watermarking, which proceeds in three steps: (1) select a mature layer and an amateur layer among the decoding layers, (2) identify a pivot layer using a watermark-related question to assess whether the layer is visually well-grounded, and (3) apply tri-layer contrastive decoding to generate the final output. Experiments on public benchmarks such as POPE, MME and AMBER demonstrate that our method achieves state-of-the-art performance in reducing hallucinations in LVLMs and generates more visually grounded responses.","short_abstract":"Large Vision-Language Models (LVLMs) have recently shown promising results on various multimodal tasks, even achieving human-comparable performance in certain cases. Nevertheless, LVLMs remain prone to hallucinations -- they often rely heavily on a single modality or memorize training data without properly grounding th...","url_abs":"https://arxiv.org/abs/2510.14304","url_pdf":"https://arxiv.org/pdf/2510.14304v1","authors":"[\"Kyungryul Back\",\"Seongbeom Park\",\"Milim Kim\",\"Mincheol Kwon\",\"SangHyeok Lee\",\"Hyunyoung Lee\",\"Junhee Cho\",\"Seunghyun Park\",\"Jinkyu Kim\"]","published":"2025-10-16T04:58:45Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Language Model\"]","has_code":false}