思考前先批判：通过理由增强的指令微调缓解幻觉
Critique Before Thinking: Mitigating Hallucination through Rationale-Augmented Instruction Tuning
AI总结：
本文提出 Re-Critic 理由增强指令微调框架，通过视觉理由合成与上下文自我批评偏好优化，提升 LVLM 回答的视觉依据并改善多模态推理。
摘要：
尽管多模态推理任务取得了显著进展，但现有的大型视觉-语言模型（LVLMs）在解释相关图像时，仍容易产生缺乏视觉依据的回答。相比之下，人类在开始学习新知识时，往往依赖一套基本的预习原则：回顾大纲以掌握核心概念，总结要点以引导注意力并增进理解。然而，当前的指令微调过程明显缺少这类准备性动作。本文提出 Re-Critic，这是一个易于扩展的理由增强框架，旨在将基本规则与思维链（CoT）作为桥梁纳入其中，以增强推理能力。具体而言，Re-Critic 开发了一个视觉理由合成器，可通过理由解释对原始指令进行可扩展的增强。为了探究更具上下文依据的回答，Re-Critic 采用上下文内自我批评机制来选择响应用于偏好微调的回答对。实验表明，使用我们的理由增强数据集微调的模型，其收益不仅限于幻觉相关任务，还扩展到更广泛的多模态推理任务。
英文摘要：
Despite significant advancements in multimodal reasoning tasks, existing Large Vision-Language Models (LVLMs) are prone to producing visually ungrounded responses when interpreting associated images. In contrast, when humans embark on learning new knowledge, they often rely on a set of fundamental pre-study principles: reviewing outlines to grasp core concepts, summarizing key points to guide their focus and enhance understanding. However, such preparatory actions are notably absent in the current instruction tuning processes. This paper presents Re-Critic, an easily scalable rationale-augmented framework designed to incorporate fundamental rules and chain-of-thought (CoT) as a bridge to enhance reasoning abilities. Specifically, Re-Critic develops a visual rationale synthesizer that scalably augments raw instructions with rationale explanation. To probe more contextually grounded responses, Re-Critic employs an in-context self-critic mechanism to select response pairs for preference tuning. Experiments demonstrate that models fine-tuned with our rationale-augmented dataset yield gains that extend beyond hallucination-specific tasks to broader multimodal reasoning tasks.
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。
