在线对抗幻觉的视觉-语言模型自我校准
Online Self-Calibration Against Hallucination in Vision-Language Models
AI总结：
本文提出OSCAR框架，通过蒙特卡洛树搜索和双粒度奖励机制在线校准视觉-语言模型，减少幻觉并提升多模态能力。
摘要：
大型视觉-语言模型（LVLMs）常出现幻觉，生成包含输入图像中不存在的视觉细节的描述。最近的偏好对齐方法通常依赖于从更强模型如GPT中提取的监督信息。然而，这种离线范式引入了监督-感知不匹配：学生模型被迫对超出其感知能力的细粒度细节进行对齐，学习猜测而非观察。为了获得可靠的在线学习自监督，我们识别出LVLMs中的生成-判别差距，其中模型在判别验证上比开放生成更准确。利用这一能力，我们提出在线自我校准（OSCAR）框架，通过蒙特卡洛树搜索和双粒度奖励机制构建偏好数据，并通过直接偏好优化迭代改进模型。大量实验表明，OSCAR在幻觉基准上实现最先进的性能，同时提升多模态能力。
英文摘要：
Large Vision-Language Models (LVLMs) often suffer from hallucinations, generating descriptions that include visual details absent from the input image. Recent preference alignment methods typically rely on supervision distilled from stronger models such as GPT. However, this offline paradigm introduces a Supervision-Perception Mismatch: the student model is forced to align with fine-grained details beyond its perceptual capacity, learning to guess rather than to see. To obtain reliable self-supervision for online learning, we identify a Generative-Discriminative Gap within LVLMs, where models exhibit higher accuracy on discriminative verification than open-ended generation. Leveraging this capability, we propose \textbf{O}nline \textbf{S}elf-\textbf{CA}lib\textbf{R}ation (OSCAR), a framework that integrates Monte Carlo Tree Search with a Dual-Granularity Reward Mechanism to construct preference data and iteratively refines the model via Direct Preference Optimization. Extensive experiments demonstrate that OSCAR achieves state-of-the-art performance on hallucination benchmarks while improving general multimodal capabilities.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
