Large Vision-Language Models (LVLMs) still struggle with vision hallucination, where generated responses are inconsistent with the visual input. Existing methods either rely on large-scale annotated data for fine-tuning, which incurs massive computational overhead, or employ static post-hoc strategies that overlook the dynamic nature of hallucination emergence. To address these, we introduce a new self-rewarding framework, enabling dynamic hallucination mitigation at inference time without external supervision. On the empirical side, we reveal that visual hallucination exhibits phase-wise dynamic patterns, peaking at the onset of each semantic phase. Drawing on these insights, we propose \textbf{PSRD} (\textbf{Phase-wise \textbf{S}elf-\textbf{R}eward \textbf{D}ecoding) for online hallucination correction guided by phase-wise self-reward signals. To reduce the cost of repeated self-evaluation during decoding, we distill the hallucination guidance signal from LVLMs into a lightweight reward model. The reward model subsequently provides on-the-fly guidance for targeted intervention during the decoding process, enabling precise hallucination suppression. The proposed PSRD significantly reduces the hallucination rate of LLaVA-1.5-7B by 50.0% and consistently outperforms existing post-hoc methods across five hallucination evaluation benchmarks for four LVLMs. Further analysis confirms that PSRD effectively mitigates hallucination propagation and achieves a highly controllable trade-off between strong performance and inference efficiency.

翻译：大型视觉语言模型（LVLMs）仍面临视觉幻觉问题，即生成响应与视觉输入不一致。现有方法要么依赖大规模标注数据进行微调（导致巨大计算开销），要么采用静态事后策略，忽视幻觉产生的动态演化特性。针对此问题，我们提出新型自奖励框架，可在推理阶段无需外部监督实现动态幻觉缓解。实验发现，视觉幻觉呈现相位动态模式，在每个语义相位初始阶段达到峰值。基于此洞见，我们提出\textbf{PSRD}（\textbf{相位自奖励解码}），通过相位自奖励信号实现在线幻觉纠正。为降低解码过程中重复自评估的计算成本，我们将LVLMs的幻觉引导信号蒸馏为轻量级奖励模型。该奖励模型在解码过程中实时提供定向干预指导，实现精准幻觉抑制。所提PSRD使LLaVA-1.5-7B的幻觉率降低50.0%，并在四项LVLMs的五项幻觉评估基准上持续优于现有事后方法。进一步分析证实，PSRD可有效缓解幻觉传播，并在强性能与推理效率间实现高度可控的平衡。