https://terms-en.ai-term-hub.com/zh/tags/llm-techniques/Recent content in LLM Techniques on 中文AI术语词典Hugozh-cnSat, 18 Jul 2026 11:44:45 +0000https://terms-en.ai-term-hub.com/zh/terms/chain_of_thought_prompting/Sat, 18 Jul 2026 10:59:27 +0000https://terms-en.ai-term-hub.com/zh/terms/chain_of_thought_prompting/<h2 id="definition">Definition</h2> <p>思维链（CoT）提示通过明确要求模型阐述其逐步逻辑，从而提升大型语言模型在复杂推理任务上的表现。与直接跳跃到最终答案不同，CoT引导模型展示其思考过程，这有助于解决需要多步逻辑推导的问题，如数学应用题或复杂的逻辑推理任务。</p> <h3 id="summary">Summary</h3> <p>思维链提示是一种鼓励大语言模型在生成最终答案之前，先生成中间推理步骤的技术。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>逐步推理</li> <li>少样本学习</li> <li>逻辑演绎</li> <li>中间步骤</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>解决数学应用题</li> <li>复杂逻辑推理任务</li> <li>调试代码生成错误</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span>prompt <span style="color:#f92672">=</span> <span style="color:#e6db74">&#34;Q: Roger has 5 tennis balls. He buys 2 more cans of tennis balls. If each can has 3 balls, how many does he have?</span><span style="color:#ae81ff">\n</span><span style="color:#e6db74">A: Roger started with 5 balls. 2 cans of 3 balls each is 6 balls. 5 + 6 = 11. The answer is 11.&#34;</span> </span></span><span style="display:flex;"><span>print(prompt) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/zero-shot-prompting-%E9%9B%B6%E6%A0%B7%E6%9C%AC%E6%8F%90%E7%A4%BA/">Zero-Shot Prompting (零样本提示)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/few-shot-prompting-%E5%B0%91%E6%A0%B7%E6%9C%AC%E6%8F%90%E7%A4%BA/">Few-Shot Prompting (少样本提示)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/self-consistency-%E8%87%AA%E6%B4%BD%E6%80%A7/">Self-Consistency (自洽性)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/reasoning-%E6%8E%A8%E7%90%86/">Reasoning (推理)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/in_context_learning/Sat, 18 Jul 2026 07:44:46 +0000https://terms-en.ai-term-hub.com/zh/terms/in_context_learning/<h2 id="definition">Definition</h2> <p>上下文学习（ICL）允许大型语言模型在不更新权重的情况下适应新任务。通过在提示上下文中提供输入-输出对，模型可以推断出模式并执行相应任务。</p> <h3 id="summary">Summary</h3> <p>一种模型通过观察提示中提供的示例来执行任务的技术。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>少样本学习</li> <li>零样本</li> <li>提示设计</li> <li>无权重适配</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>快速测试模型在新数据集上的能力</li> <li>对话代理中的动态任务切换</li> <li>无需重新训练即可原型化AI应用</li> </ul> <h2 id="code-example">Code Example</h2> <div class="highlight"><pre tabindex="0" style="color:#f8f8f2;background-color:#272822;-moz-tab-size:4;-o-tab-size:4;tab-size:4;"><code class="language-python" data-lang="python"><span style="display:flex;"><span>prompt <span style="color:#f92672">=</span> <span style="color:#e6db74">&#34;Translate English to French:</span><span style="color:#ae81ff">\n</span><span style="color:#e6db74">English: Hello</span><span style="color:#ae81ff">\n</span><span style="color:#e6db74">French: Bonjour</span><span style="color:#ae81ff">\n</span><span style="color:#e6db74">English: Cat</span><span style="color:#ae81ff">\n</span><span style="color:#e6db74">French:&#34;</span> </span></span><span style="display:flex;"><span>response <span style="color:#f92672">=</span> model<span style="color:#f92672">.</span>generate(prompt) </span></span></code></pre></div><h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%8F%90%E7%A4%BA%E5%B7%A5%E7%A8%8B/">提示工程</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%B0%91%E6%A0%B7%E6%9C%AC/">少样本</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E9%9B%B6%E6%A0%B7%E6%9C%AC/">零样本</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%85%83%E5%AD%A6%E4%B9%A0/">元学习</a></li> </ul>