研究人员开发了一个名为“意图投影”（Intent Projection）的新框架，以改进大型视觉语言模型（LVLMs）对模因等多模态内容的语用含义的理解。该方法将图像和文本的字面描述与作者的预期沟通分离开来。该框架通过修改模型的表示、输出和目标函数来实现这一点，从而在各种基准测试中取得更好的性能，尤其是在处理复杂或讽刺性帖子时。 AI
影响 增强了AI理解细微交流的能力，有望改善人机在社交情境中的互动。
排序理由 该集群包含一篇详细介绍新研究框架及其在基准测试中表现的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2606.03604v1 Announce Type: new Abstract: When asked what a meme or sarcastic post means, Large Vision Language Models (LVLMs) tend to describe what the image shows rather than what the author is trying to communicate. Standard instruction tuning entangles a post's literal …
When asked what a meme or sarcastic post means, Large Vision Language Models (LVLMs) tend to describe what the image shows rather than what the author is trying to communicate. Standard instruction tuning entangles a post's literal content with its pragmatic meaning, letting surf…