研究人员推出 VCap，一种用于视觉字幕模型的新型奖励机制，旨在提高事实准确性和覆盖率。VCap 采用证人-仲裁者方法，将参考字幕与视觉信号配对以验证事实一致性。即使参考字幕不完美，该方法也能实现有效学习，从而在强化学习训练中实现弱到强的泛化。实验表明，使用 VCap 训练的 8B 模型在各种字幕基准测试中均优于最先进的模型，并展示了改进的感知能力。 AI
影响 VCap 的事实验证方法可能带来更可靠、更准确的图像和视频描述，影响依赖多模态理解的应用。
排序理由 该集群描述了一篇关于视觉字幕新方法的最新研究论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2605.28023v1 Announce Type: cross Abstract: Visual captioning requires models to capture visual content faithfully while minimizing both omission and hallucination. As the dominant paradigm for captioning, MLLMs have achieved strong performance through scaling and high-qual…
Visual captioning requires models to capture visual content faithfully while minimizing both omission and hallucination. As the dominant paradigm for captioning, MLLMs have achieved strong performance through scaling and high-quality data. Recently, RL has emerged as a key route …