mcpc.site把论文句子读懂
🔥 1 天
← 全部句式 g23 · L1
However, X ... / X (e.g., Y) ... / X (i.e., Y) ...

不过…… / 例如 Y / 也就是说 Y(插在句中的补充)

插入语

为什么这么写

读懂 + 会看

读法:括号里的 (e.g., ...) 是举例,(i.e., ...) 是「换句话说、也就是」,两者都可以先整块跳过,主干不受影响。判断口诀:e.g. = for example,后面给的是例子之一;i.e. = that is,前后可以画等号,是等价改写。However 是句首的转向灯,看到它就知道下面要跟「但是」,前面那句大概率在讲别人怎么做。为什么作者这么写:论文术语密集,用括号就地解释可以避免另起一句打断节奏;However 则把「前人做法 → 但有问题」的转折一步到位。读的时候先把括号拿掉读一遍主干,再决定要不要回头看括号——很多时候不看也不影响理解。

论文里的例子

2 条

However, most existing geometry reconstruction methods (e.g., DVGT) rely on computationally expensive batch processing of multi-frame inputs and cannot be applied to online planning.

DVGT-2 (arXiv 2604.00813) · Abstract

Our method jointly performs spatial understanding, 3D perception (i.e., 3D occupancy), prediction (i.e., occupancy flow), and planning (i.e., action outputs) in parallel through end-to-end optimization.

DrivePI (arXiv 2512.12799) · Abstract

练这几句

拆句练习
011 While large models still achieve stronger accuracy on some datasets, such as Waymo and PandaSet, LiAuto-GeoX maintains comparable point-cloud reconstruction quality under a much smaller model scale. → 012 However, most existing geometry reconstruction methods (e.g., DVGT) rely on computationally expensive batch processing of multi-frame inputs and cannot be applied to online planning. → 029 However, existing LLM-based methods struggle to directly output fine-grained perception results (e.g., 3D occupancy and occupancy flow) through next-token prediction, which are readily achievable by VA approaches. → 042 This significant performance gain, especially the dramatic improvement in safety, strongly indicates that by introducing causal intervention, the model learns to better understand scene dynamics and make more reliable decisions. →

怎么用

先写一句「现有方法依赖昂贵的标注」,再用 (e.g., ...) 在括号里补两个标注类型的例子,最后用 However 开头接一句转折。