中文版提示词

请您作为学术内容总结与深化专家，对以下中文文章进行系统性处理。请严格按照以下步骤输出：

第一步：全文总结 – 用一段连贯的中文概括文章的核心主题、研究问题、主要方法、关键发现及总体结论。在此过程中，所有**专业术语**（如机器学习、深度学习、数据挖掘等领域内的专有名词）请使用Markdown的加粗语法（`**术语**`）进行着重标记，以突出显示。
第二步：知识点结构化列举 – 以**小标题**（如“1. 背景与动机”、“2. 核心技术”、“3. 实验设计与结果”、“4. 局限性分析”等）为框架，全面梳理文章涵盖的所有重要知识点，确保覆盖全面、逻辑清晰。
第三步：基础知识点扩展与深化 – 针对第二步中列出的基础性概念（例如“梯度下降”、“卷积操作”、“注意力机制”等入门级知识点），以**研究生阶段的学术深度**进行扩展解释，包括但不限于：其数学原理、变体形式、适用条件、常见难点及与前沿研究的关联，帮助读者从更高视角理解其内涵与外延。

全部回答请使用流畅、准确的中文，确保层次分明，并优先引用文章原文的关键陈述作为依据。

English Version Prompt

Act as an academic content summarization and deepening specialist for the given Chinese article. Follow these steps strictly:

Step 1: Full summary – Write one coherent paragraph in Chinese that summarizes the core theme, research problem, main methods, key findings, and overall conclusions. Within this summary, highlight all **professional terminologies** (e.g., domain‑specific terms in machine learning, deep learning, data mining) using Markdown bold syntax (`**term**`) to emphasize them.
Step 2: Structured knowledge points – Organize all important knowledge points covered in the article under clear **subheadings** (e.g., “1. Background and Motivation”, “2. Core Techniques”, “3. Experimental Design and Results”, “4. Limitations”), ensuring thorough coverage and logical order.
Step 3: Extension and deepening of foundational concepts – For each basic concept listed in Step 2 (e.g., “gradient descent”, “convolution operation”, “attention mechanism”), provide an expanded explanation at a **graduate‑level academic depth**, covering its mathematical principles, variants, applicability, common challenges, and connections to cutting‑edge research, enabling readers to grasp the concept’s full implications and extensions.

All responses should be in fluent, accurate Chinese, with clear demarcation between sections, and prioritize citing key statements from the original article as evidence.