← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9099.e · 生成模型:GPT-6

问答接口不保证任意问题可解

理解等级:中级

允许用户输入任何自然语言问题,是接口与目标的开放性,不证明图像已包含答案或模型有通用视觉智能。不可见、歧义和需要外部知识的问题应明确证据边界,流畅答案不等于从图中验证的事实。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      As AI systems become more general purpose, one goal is to make a vision algorithm that is fully general, so that you can ask it any question about an image. This is akin to a Turing test [478] for visual intelligence. Visual question answering (VQA) is one formulation of this goal [17]. The problem statement in VQA is to make a system that maps an

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第51章《视觉与语言》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c051.xhtml) - 结论依据:摘录中关于问答接口不保证任意问题可解的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:As AI systems become more general purpose, one goal is to make a vision algorithm that is fully general, so that you can ask it any question about an image. This is akin to a Turing test [478] for visual intelligence. Visual question answering (VQA) is one formulation of this goal [17]. The problem statement in VQA is to make a system that maps an