ReflectCAP: 基于反思记忆的详细图像描述
ReflectCAP: Detailed Image Captioning with Reflective Memory
AI总结：
ReflectCAP通过反思笔记指导生成详细且准确的图像描述，平衡事实性和覆盖性，优于现有方法。
摘要：
详细图像描述需要事实基础和精细覆盖，但现有方法难以兼顾。ReflectCAP通过多代理流程分析目标大视觉-语言模型的幻觉和遗漏，生成结构化反思笔记。在推理时，这些笔记指导描述模型避免和关注内容，生成同时提升事实性和覆盖性的详细描述。在8种LVLM上应用后，ReflectCAP达到事实性和覆盖性的帕累托前沿，并在CapArena-Auto中表现优异。此外，ReflectCAP在描述质量和计算成本之间提供更优的权衡，优于模型扩展或现有多代理流程。
英文摘要：
Detailed image captioning demands both factual grounding and fine-grained coverage, yet existing methods have struggled to achieve them simultaneously. We address this tension with Reflective Note-Guided Captioning (ReflectCAP), where a multi-agent pipeline analyzes what the target large vision-language model (LVLM) consistently hallucinates and what it systematically overlooks, distilling these patterns into reusable guidelines called Structured Reflection Notes. At inference time, these notes steer the captioning model along both axes -- what to avoid and what to attend to -- yielding detailed captions that jointly improve factuality and coverage. Applying this method to 8 LVLMs spanning the GPT-4.1 family, Qwen series, and InternVL variants, ReflectCAP reaches the Pareto frontier of the trade-off between factuality and coverage, and delivers substantial gains on CapArena-Auto, where generated captions are judged head-to-head against strong reference models. Moreover, ReflectCAP offers a more favorable trade-off between caption quality and compute cost than model scaling or existing multi-agent pipelines, which incur 21--36\% greater overhead. This makes high-quality detailed captioning viable under real-world cost and latency constraints.
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。
