https://terms-en.ai-term-hub.com/zh/tags/efficiency/Recent content in Efficiency 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/unified_model/Sat, 18 Jul 2026 11:37:26 +0000https://terms-en.ai-term-hub.com/zh/terms/unified_model/<h2 id="definition">Definition</h2> <p>统一模型是指能够执行各种不同任务（如文本生成、图像识别和代码合成）的人工智能系统，而无需依赖单独的专业模型。</p> <h3 id="summary">Summary</h3> <p>旨在单个框架内处理多项任务或多模态数据的AI架构。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>多模态学习</li> <li>任务无关性</li> <li>架构效率</li> <li>互操作性</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>构建多功能AI助手</li> <li>降低基础设施成本</li> <li>简化开发流水线</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/multimodal-ai-%E5%A4%9A%E6%A8%A1%E6%80%81ai/">Multimodal AI (多模态AI)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/foundation-models-%E5%9F%BA%E7%A1%80%E6%A8%A1%E5%9E%8B/">Foundation Models (基础模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/modular-ai-%E6%A8%A1%E5%9D%97%E5%8C%96ai/">Modular AI (模块化AI)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/general-purpose-ai-%E9%80%9A%E7%94%A8%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD/">General Purpose AI (通用人工智能)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/robotic_process_automation/Sat, 18 Jul 2026 11:32:26 +0000https://terms-en.ai-term-hub.com/zh/terms/robotic_process_automation/<h2 id="definition">Definition</h2> <p>机器人流程自动化（RPA）采用软件机器人（通常增强有AI功能）来模拟人类与数字系统的交互。它用于简化工作流程，如数据录入、发票处理等，从而提高效率并减少错误。</p> <h3 id="summary">Summary</h3> <p>使用软件机器人自动化传统上由人类在业务流程中执行的高容量、重复性数字任务。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>软件机器人</li> <li>工作流自动化</li> <li>基于规则的处理</li> <li>数字劳动力</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>发票处理</li> <li>客户入职流程</li> <li>数据迁移任务</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/business_process_automation-%E4%B8%9A%E5%8A%A1%E6%B5%81%E7%A8%8B%E8%87%AA%E5%8A%A8%E5%8C%96/">business_process_automation (业务流程自动化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/cognitive_computing-%E8%AE%A4%E7%9F%A5%E8%AE%A1%E7%AE%97/">cognitive_computing (认知计算)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/workflow_engine-%E5%B7%A5%E4%BD%9C%E6%B5%81%E5%BC%95%E6%93%8E/">workflow_engine (工作流引擎)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/digital_worker-%E6%95%B0%E5%AD%97%E5%91%98%E5%B7%A5/">digital_worker (数字员工)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/qwen3_5_moe/Sat, 18 Jul 2026 11:31:37 +0000https://terms-en.ai-term-hub.com/zh/terms/qwen3_5_moe/<h2 id="definition">Definition</h2> <p>该术语指代 Qwen 家族中的一种专用架构，可能利用了混合专家（Mixture of Experts, MoE）设计。在此类模型中，仅激活神经网络参数（即“专家”）的一个子集，从而在保持性能的同时显著降低计算开销和内存需求。</p> <h3 id="summary">Summary</h3> <p>一种假设或未来的稀疏混合专家（MoE）变体，属于 Qwen3 系列，专为高效率设计。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>混合专家</li> <li>稀疏激活</li> <li>推理效率</li> <li>大语言模型</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>高吞吐量 API 服务</li> <li>边缘设备部署</li> <li>具有成本效益的扩展</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/sparse-attention-%E7%A8%80%E7%96%8F%E6%B3%A8%E6%84%8F%E5%8A%9B%E6%9C%BA%E5%88%B6/">Sparse Attention (稀疏注意力机制)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/model-quantization-%E6%A8%A1%E5%9E%8B%E9%87%8F%E5%8C%96/">Model Quantization (模型量化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transformer-architecture-transformer-%E6%9E%B6%E6%9E%84/">Transformer Architecture (Transformer 架构)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/quantized/Sat, 18 Jul 2026 11:31:12 +0000https://terms-en.ai-term-hub.com/zh/terms/quantized/<h2 id="definition">Definition</h2> <p>量化是一种模型优化技术，它降低了机器学习模型参数的数值精度，通常将 32 位浮点数转换为 8 位整数。这种方法可以在保持模型性能基本不变的情况下，显著减少内存占用并加速推理速度。</p> <h3 id="summary">Summary</h3> <p>量化后的指神经网络模型中权重和激活值以较低精度的数字表示，以减少模型大小和推理延迟。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>模型压缩</li> <li>低精度算术</li> <li>边缘部署</li> <li>推理优化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>移动 AI 应用</li> <li>物联网设备集成</li> <li>实时视频处理</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%89%AA%E6%9E%9D-pruning/">剪枝 (Pruning)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%9F%A5%E8%AF%86%E8%92%B8%E9%A6%8F-distillation/">知识蒸馏 (Distillation)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/tensorflow-lite/">TensorFlow Lite</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/pruning/Sat, 18 Jul 2026 11:30:52 +0000https://terms-en.ai-term-hub.com/zh/terms/pruning/<h2 id="definition">Definition</h2> <p>剪枝涉及识别并消除神经网络中对输出准确率贡献最小的神经元、连接或滤波器。通过移除这些冗余元素，模型变得更加紧凑，从而降低存储需求和计算开销，同时尽量保持原有的模型性能。</p> <h3 id="summary">Summary</h3> <p>一种模型压缩技术，通过移除冗余或重要性较低的参数来减小模型体积并提高推理速度。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>模型压缩</li> <li>冗余移除</li> <li>推理加速</li> <li>稀疏性</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>移动端AI部署</li> <li>边缘计算优化</li> <li>降低云端推理成本</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E9%87%8F%E5%8C%96-quantization/">量化 (Quantization)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%9F%A5%E8%AF%86%E8%92%B8%E9%A6%8F-knowledge-distillation/">知识蒸馏 (Knowledge Distillation)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%A8%A1%E5%9E%8B%E5%8E%8B%E7%BC%A9-model-compression/">模型压缩 (Model Compression)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%A8%80%E7%96%8F%E7%BD%91%E7%BB%9C-sparse-networks/">稀疏网络 (Sparse Networks)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/prompt_tuning/Sat, 18 Jul 2026 11:30:52 +0000https://terms-en.ai-term-hub.com/zh/terms/prompt_tuning/<h2 id="definition">Definition</h2> <p>提示微调涉及在预训练语言模型的输入层添加可训练的软提示（连续向量），同时保持底层模型参数冻结。这种方法允许在几乎不增加计算成本的情况下，使模型适应特定任务或领域，特别适用于资源受限的场景。</p> <h3 id="summary">Summary</h3> <p>一种参数高效的微调方法，通过优化连续的输入嵌入（embeddings）而非更新整个模型权重来调整模型。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>软提示 (Soft Prompts)</li> <li>参数效率</li> <li>冻结权重</li> <li>少样本学习 (Few-shot Learning)</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>针对特定领域适配大语言模型</li> <li>低资源环境下的微调</li> <li>多任务学习优化</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/peft-parameter-efficient-fine-tuning-%E5%8F%82%E6%95%B0%E9%AB%98%E6%95%88%E5%BE%AE%E8%B0%83/">PEFT (Parameter-Efficient Fine-Tuning, 参数高效微调)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/lora-low-rank-adaptation-%E4%BD%8E%E7%A7%A9%E8%87%AA%E9%80%82%E5%BA%94/">LoRA (Low-Rank Adaptation, 低秩自适应)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E4%B8%8A%E4%B8%8B%E6%96%87%E5%AD%A6%E4%B9%A0-in-context-learning/">上下文学习 (In-context Learning)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%BE%AE%E8%B0%83-fine-tuning/">微调 (Fine-tuning)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/proactive_learning/Sat, 18 Jul 2026 11:30:30 +0000https://terms-en.ai-term-hub.com/zh/terms/proactive_learning/<h2 id="definition">Definition</h2> <p>在主动学习中，AI系统确定哪些样本最能降低不确定性或提高模型性能，通常通过主动学习或探索策略实现。这与被动接收数据的传统监督学习形成对比，显著提高了数据效率。</p> <h3 id="summary">Summary</h3> <p>一种学习策略，智能体主动选择提供信息量大的数据点进行查询或探索，而非被动接收。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>主动学习</li> <li>查询策略</li> <li>不确定性采样</li> <li>数据效率</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>高效标注医疗记录</li> <li>自主机器人探索</li> <li>有限标注的情感分析</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/active-learning-%E4%B8%BB%E5%8A%A8%E5%AD%A6%E4%B9%A0/">Active Learning (主动学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/exploration-vs-exploitation-%E6%8E%A2%E7%B4%A2%E4%B8%8E%E5%88%A9%E7%94%A8/">Exploration vs Exploitation (探索与利用)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/sample-efficiency-%E6%A0%B7%E6%9C%AC%E6%95%88%E7%8E%87/">Sample Efficiency (样本效率)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/human-in-the-loop-%E4%BA%BA%E5%9C%A8%E5%9B%9E%E8%B7%AF/">Human-in-the-loop (人在回路)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/prefix_tuning/Sat, 18 Jul 2026 11:30:18 +0000https://terms-en.ai-term-hub.com/zh/terms/prefix_tuning/<h2 id="definition">Definition</h2> <p>前缀微调是一种用于预训练变换器的参数高效适应技术。与更新所有模型权重不同，它在输入序列前prepend（前置）一系列可训练的连续向量（即前缀），从而冻结主干网络并仅优化少量参数以适应新任务。</p> <h3 id="summary">Summary</h3> <p>一种参数高效的微调方法，通过在变换器层的输入端添加可训练的连续向量来实现适配。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>参数高效微调</li> <li>软提示</li> <li>变换器层</li> <li>冻结主干</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>少样本学习适配</li> <li>资源受限下的多任务学习</li> <li>为利基领域定制大型语言模型</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/prompt_tuning-%E6%8F%90%E7%A4%BA%E5%BE%AE%E8%B0%83/">prompt_tuning (提示微调)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/p_tuning-p-tuning/">p_tuning (P-Tuning)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/adapter_modules-%E9%80%82%E9%85%8D%E5%99%A8%E6%A8%A1%E5%9D%97/">adapter_modules (适配器模块)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/peft-%E5%8F%82%E6%95%B0%E9%AB%98%E6%95%88%E5%BE%AE%E8%B0%83%E6%8A%80%E6%9C%AF/">peft (参数高效微调技术)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/phi/Sat, 18 Jul 2026 11:29:53 +0000https://terms-en.ai-term-hub.com/zh/terms/phi/<h2 id="definition">Definition</h2> <p>Phi（基于教学学习范式的基座模型）是微软创建的一系列紧凑型大型语言模型家族。与传统在海量网络语料上训练的LLM不同，Phi主要通过合成数据进行训练，旨在以更小的参数量实现更高的知识密度和逻辑推理能力。</p> <h3 id="summary">Summary</h3> <p>由微软研究院开发的一系列小型但高效的大型语言模型，侧重于知识密度和推理能力。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>小型语言模型</li> <li>合成数据</li> <li>推理能力</li> <li>知识密度</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>边缘设备AI部署</li> <li>复杂逻辑推理任务</li> <li>高性价比API服务</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/large_language_model-%E5%A4%A7%E5%9E%8B%E8%AF%AD%E8%A8%80%E6%A8%A1%E5%9E%8B/">large_language_model (大型语言模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/microsoft_research-%E5%BE%AE%E8%BD%AF%E7%A0%94%E7%A9%B6%E9%99%A2/">microsoft_research (微软研究院)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/synthetic_data_generation-%E5%90%88%E6%88%90%E6%95%B0%E6%8D%AE%E7%94%9F%E6%88%90/">synthetic_data_generation (合成数据生成)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/p_tuning/Sat, 18 Jul 2026 11:29:14 +0000https://terms-en.ai-term-hub.com/zh/terms/p_tuning/<h2 id="definition">Definition</h2> <p>P-Tuning（提示微调）是一种旨在以最低计算成本将大型预训练语言模型适配到特定下游任务的技术。与微调所有模型参数不同，P-Tuning 仅优化少量可学习的提示向量，同时保持预训练模型的权重冻结，从而显著降低资源消耗并防止灾难性遗忘。</p> <h3 id="summary">Summary</h3> <p>P-Tuning 是一种参数高效微调方法，它通过优化连续的提示嵌入（prompt embeddings）来适应任务，而不是更新整个预训练模型的权重。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>参数高效微调</li> <li>虚拟令牌</li> <li>冻结权重</li> <li>嵌入优化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>少样本学习适配</li> <li>资源受限环境</li> <li>大语言模型应用的快速原型开发</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/lora-%E4%BD%8E%E7%A7%A9%E9%80%82%E5%BA%94/">LoRA (低秩适应)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/adapter-modules-%E9%80%82%E9%85%8D%E5%99%A8%E6%A8%A1%E5%9D%97/">Adapter Modules (适配器模块)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/prompt-engineering-%E6%8F%90%E7%A4%BA%E5%B7%A5%E7%A8%8B/">Prompt Engineering (提示工程)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transfer-learning-%E8%BF%81%E7%A7%BB%E5%AD%A6%E4%B9%A0/">Transfer Learning (迁移学习)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/multitask_optimization/Sat, 18 Jul 2026 11:27:47 +0000https://terms-en.ai-term-hub.com/zh/terms/multitask_optimization/<h2 id="definition">Definition</h2> <p>多任务优化涉及训练单个模型同时处理几个不同但相关的任务。通过在任务间共享中间表示，模型可以学习到更通用的特征，从而提高泛化能力和效率。</p> <h3 id="summary">Summary</h3> <p>一种训练策略，使模型能够同时优化执行多个相关任务。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>共享表示</li> <li>任务特定头部</li> <li>梯度平衡</li> <li>迁移学习</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>联合目标检测与分割</li> <li>多标签分类</li> <li>语音识别与语言建模</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transfer_learning-%E8%BF%81%E7%A7%BB%E5%AD%A6%E4%B9%A0/">transfer_learning (迁移学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/multi_label_classification-%E5%A4%9A%E6%A0%87%E7%AD%BE%E5%88%86%E7%B1%BB/">multi_label_classification (多标签分类)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/shared_layers-%E5%85%B1%E4%BA%AB%E5%B1%82/">shared_layers (共享层)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/gradient_accumulation-%E6%A2%AF%E5%BA%A6%E7%B4%AF%E7%A7%AF/">gradient_accumulation (梯度累积)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/multi_task_learning/Sat, 18 Jul 2026 11:26:49 +0000https://terms-en.ai-term-hub.com/zh/terms/multi_task_learning/<h2 id="definition">Definition</h2> <p>该技术利用相关任务之间共享的归纳偏置来提高学习效率 and 性能。通过训练单个模型同时执行多项任务，模型能够学习到更具通用性的特征表示，从而提升整体表现。</p> <h3 id="summary">Summary</h3> <p>多任务学习是一种机器学习范式，其中模型同时训练多个相关任务以提高泛化能力。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>共享表示</li> <li>归纳偏置</li> <li>任务干扰</li> <li>泛化能力</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>自然语言处理（如命名实体识别和词性标注）</li> <li>计算机视觉（如目标检测和分割）</li> <li>语音识别与合成</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transfer-learning-%E8%BF%81%E7%A7%BB%E5%AD%A6%E4%B9%A0/">Transfer Learning (迁移学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/joint-training-%E8%81%94%E5%90%88%E8%AE%AD%E7%BB%83/">Joint Training (联合训练)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/neural-architecture-search-%E7%A5%9E%E7%BB%8F%E6%9E%B6%E6%9E%84%E6%90%9C%E7%B4%A2/">Neural Architecture Search (神经架构搜索)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/regularization-%E6%AD%A3%E5%88%99%E5%8C%96/">Regularization (正则化)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/moe/Sat, 18 Jul 2026 11:26:37 +0000https://terms-en.ai-term-hub.com/zh/terms/moe/<h2 id="definition">Definition</h2> <p>混合专家（MoE）是一种旨在提高效率和可扩展性的机器学习架构。MoE 不使用单个大型模型处理所有任务，而是采用多个较小的“专家”网络。</p> <h3 id="summary">Summary</h3> <p>一种架构模式，通过门控机制结合多个专用神经网络（专家）来处理输入。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>稀疏激活</li> <li>门控网络</li> <li>专家专业化</li> <li>可扩展性</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>高效训练大型语言模型</li> <li>降低大型模型的推理延迟</li> <li>在多模态系统中处理多样化的输入类型</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%A8%80%E7%96%8F-transformer-sparse-transformers/">稀疏 Transformer (Sparse Transformers)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%9D%A1%E4%BB%B6%E8%AE%A1%E7%AE%97-conditional-computation/">条件计算 (Conditional Computation)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%A4%A7%E5%9E%8B%E8%AF%AD%E8%A8%80%E6%A8%A1%E5%9E%8B-large-language-models/">大型语言模型 (Large Language Models)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%A5%9E%E7%BB%8F%E6%9E%B6%E6%9E%84%E6%90%9C%E7%B4%A2-neural-architecture-search/">神经架构搜索 (Neural Architecture Search)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/model_compression/Sat, 18 Jul 2026 11:26:24 +0000https://terms-en.ai-term-hub.com/zh/terms/model_compression/<h2 id="definition">Definition</h2> <p>该类别包括剪枝、量化和知识蒸馏等方法，旨在缩小模型规模的同时保持性能。这对于部署复杂的人工智能模型至关重要，尤其是在资源受限的环境中。</p> <h3 id="summary">Summary</h3> <p>模型压缩是指减少机器学习模型体积和计算需求的技术。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>量化</li> <li>剪枝</li> <li>知识蒸馏</li> <li>推理速度</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>在移动设备上部署模型</li> <li>降低云端推理成本</li> <li>加速实时视频处理</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><span style="color:#f92672">import</span> torch.quantization <span style="color:#66d9ef">as</span> quant </span></span><span style="display:flex;"><span>model <span style="color:#f92672">=</span> quant<span style="color:#f92672">.</span>quantize_dynamic(model, {torch<span style="color:#f92672">.</span>nn<span style="color:#f92672">.</span>Linear}, dtype<span style="color:#f92672">=</span>torch<span style="color:#f92672">.</span>qint8) </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/quantization-%E9%87%8F%E5%8C%96/">Quantization (量化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/pruning-%E5%89%AA%E6%9E%9D/">Pruning (剪枝)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/distillation-%E8%92%B8%E9%A6%8F/">Distillation (蒸馏)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/edge-ai-%E8%BE%B9%E7%BC%98%E4%BA%BA%E5%B7%A5%E6%99%BA%E8%83%BD/">Edge AI (边缘人工智能)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/mixtral/Sat, 18 Jul 2026 11:26:13 +0000https://terms-en.ai-term-hub.com/zh/terms/mixtral/<h2 id="definition">Definition</h2> <p>Mixtral是一款开创性的开源权重LLM，利用稀疏混合专家（MoE）架构。与每个令牌都使用所有参数的密集模型不同，Mixtral将每个令牌路由通</p> <h3 id="summary">Summary</h3> <p>由Mistral AI开发的稀疏混合专家（MoE）大型语言模型，每个令牌仅激活部分参数。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>稀疏MoE</li> <li>专家</li> <li>路由</li> <li>效率</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>高吞吐量推理</li> <li>复杂推理任务</li> <li>对成本敏感的生产部署</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mistral-mistral%E6%A8%A1%E5%9E%8B/">Mistral (Mistral模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mixture-of-experts-%E6%B7%B7%E5%90%88%E4%B8%93%E5%AE%B6%E6%A8%A1%E5%9E%8B/">Mixture of Experts (混合专家模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/llama-%E5%A4%A7%E5%9E%8B%E8%AF%AD%E8%A8%80%E6%A8%A1%E5%9E%8B%E6%9E%B6%E6%9E%84/">LLaMA (大型语言模型架构)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/sparsity-%E7%A8%80%E7%96%8F%E6%80%A7/">Sparsity (稀疏性)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/deepseek_v3/Sat, 18 Jul 2026 11:14:03 +0000https://terms-en.ai-term-hub.com/zh/terms/deepseek_v3/<h2 id="definition">Definition</h2> <p>DeepSeek V3是DeepSeek模型家族中的先进迭代版本，其特点是通过混合专家路由机制，在推理过程中仅激活少量参数子集。这种稀疏激活架构使得模型在保持强大推理能力的同时，大幅降低了计算和训练成本。</p> <h3 id="summary">Summary</h3> <p>DeepSeek V3是一款高效的混合专家（MoE）语言模型，以显著降低的训练成本实现了最先进的性能。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>混合专家模型</li> <li>稀疏激活</li> <li>训练效率</li> <li>推理能力</li> <li>混合优化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>复杂数学问题解决</li> <li>高级代码重构</li> <li>高风险逻辑推理任务</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/deepseek-deepseek/">DeepSeek (DeepSeek)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/moe-mixture-of-experts-%E6%B7%B7%E5%90%88%E4%B8%93%E5%AE%B6%E6%A8%A1%E5%9E%8B/">MoE (Mixture of Experts, 混合专家模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%A8%80%E7%96%8F%E6%A8%A1%E5%9E%8B-sparse-models/">稀疏模型 (Sparse Models)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/llm%E5%9F%BA%E5%87%86%E6%B5%8B%E8%AF%95-llm-benchmarking/">LLM基准测试 (LLM Benchmarking)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/business_process_automation/Sat, 18 Jul 2026 11:09:30 +0000https://terms-en.ai-term-hub.com/zh/terms/business_process_automation/<h2 id="definition">Definition</h2> <p>业务流程自动化 (BPA) 涉及利用软件和人工智能来简化复杂的业务工作流程。与仅处理基于规则任务的简单机器人流程自动化 (RPA) 不同，BPA 通常……</p> <h3 id="summary">Summary</h3> <p>使用技术来执行业务中的重复性任务或流程，从而取代人工努力。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>工作流优化</li> <li>RPA (机器人流程自动化)</li> <li>人工智能集成</li> <li>运营效率</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>自动发票处理</li> <li>客户服务聊天机器人</li> <li>人力资源入职工作流</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/rpa-%E6%9C%BA%E5%99%A8%E4%BA%BA%E6%B5%81%E7%A8%8B%E8%87%AA%E5%8A%A8%E5%8C%96/">RPA (机器人流程自动化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/digital-transformation-%E6%95%B0%E5%AD%97%E5%8C%96%E8%BD%AC%E5%9E%8B/">Digital Transformation (数字化转型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/workflow-engine-%E5%B7%A5%E4%BD%9C%E6%B5%81%E5%BC%95%E6%93%8E/">Workflow Engine (工作流引擎)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/enterprise-software-%E4%BC%81%E4%B8%9A%E8%BD%AF%E4%BB%B6/">Enterprise Software (企业软件)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/batch_processing/Sat, 18 Jul 2026 11:08:51 +0000https://terms-en.ai-term-hub.com/zh/terms/batch_processing/<h2 id="definition">Definition</h2> <p>批处理涉及在执行计算或模型推理之前，将数据输入聚合为一个组（即批次）。与实时流处理相比，这种方法允许对数据进行批量操作，从而提高效率。</p> <h3 id="summary">Summary</h3> <p>一种计算方法，数据随时间收集并以组（而非单个）为单位进行处理。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>吞吐量优化</li> <li>资源利用率</li> <li>离线计算</li> <li>分组执行</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>在历史数据集上训练大型神经网络</li> <li>数据仓库中的计划ETL作业</li> <li>夜间报告生成</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/streaming_processing-%E6%B5%81%E5%A4%84%E7%90%86/">streaming_processing (流处理)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/throughput-%E5%90%9E%E5%90%90%E9%87%8F/">throughput (吞吐量)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/data_pipeline-%E6%95%B0%E6%8D%AE%E7%AE%A1%E9%81%93/">data_pipeline (数据管道)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/offline_inference-%E7%A6%BB%E7%BA%BF%E6%8E%A8%E7%90%86/">offline_inference (离线推理)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/automated_machine_learning/Sat, 18 Jul 2026 11:07:26 +0000https://terms-en.ai-term-hub.com/zh/terms/automated_machine_learning/<h2 id="definition">Definition</h2> <p>AutoML（自动化机器学习）通过自动化数据预处理、特征工程、模型选择和超参数调整等任务，简化了ML模型的开发。它使得非专家也能使用。</p> <h3 id="summary">Summary</h3> <p>一种自动化将机器学习应用于现实世界问题端到端过程的方法论，从而减少人工工作量。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>超参数调优</li> <li>特征工程</li> <li>模型选择</li> <li>民主化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>业务分析师的快速原型设计</li> <li>优化大规模生产管道</li> <li>自动比较多种算法</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><span style="color:#f92672">from</span> auto_ml <span style="color:#f92672">import</span> Predictor </span></span><span style="display:flex;"><span>predictor <span style="color:#f92672">=</span> Predictor(type_of_estimator<span style="color:#f92672">=</span><span style="color:#e6db74">&#39;classifier&#39;</span>) </span></span><span style="display:flex;"><span>predictor<span style="color:#f92672">.</span>fit(dataframe, target<span style="color:#f92672">=</span><span style="color:#e6db74">&#39;label&#39;</span>) </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/%E8%B6%85%E5%8F%82%E6%95%B0%E4%BC%98%E5%8C%96-hyperparameter-optimization/">超参数优化 (Hyperparameter Optimization)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%A5%9E%E7%BB%8F%E6%9E%B6%E6%9E%84%E6%90%9C%E7%B4%A2-neural-architecture-search/">神经架构搜索 (Neural Architecture Search)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/mlops-mlops/">MLOps (MLOps)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%97%A0%E4%BB%A3%E7%A0%81ai-no-code-ai/">无代码AI (No-Code AI)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/automated_decision_making/Sat, 18 Jul 2026 11:07:26 +0000https://terms-en.ai-term-hub.com/zh/terms/automated_decision_making/<h2 id="definition">Definition</h2> <p>自动化决策（ADM）依靠软件系统做出以前需要人类判断的选择。常见于信用评分、内容审核和物流领域，ADM使用预定义的规则。</p> <h3 id="summary">Summary</h3> <p>算法基于数据输入独立选择行动或结果，而无需直接人工监督的过程。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>算法治理</li> <li>效率</li> <li>偏见缓解</li> <li>人在回路</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>银行信贷审批系统</li> <li>电子商务动态定价</li> <li>医疗分诊系统</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E7%AE%97%E6%B3%95%E5%81%8F%E8%A7%81-algorithmic-bias/">算法偏见 (Algorithmic Bias)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%86%B3%E7%AD%96%E6%94%AF%E6%8C%81%E7%B3%BB%E7%BB%9F-decision-support-systems/">决策支持系统 (Decision Support Systems)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%9F%BA%E4%BA%8E%E8%A7%84%E5%88%99%E7%9A%84%E7%B3%BB%E7%BB%9F-rule-based-systems/">基于规则的系统 (Rule-Based Systems)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E6%B2%BB%E7%90%86-governance/">治理 (Governance)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/active_learning/Sat, 18 Jul 2026 11:04:13 +0000https://terms-en.ai-term-hub.com/zh/terms/active_learning/<h2 id="definition">Definition</h2> <p>主动学习通过允许模型选择最具信息量的实例进行人工标记，减少了所需标记数据的数量。与被动接收随机样本不同，算法会主动识别那些最能提升模型性能的数据点进行查询。</p> <h3 id="summary">Summary</h3> <p>一种机器学习策略，算法选择性地查询用户或专家来标记新数据点，从而优化训练过程。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>查询策略</li> <li>标记效率</li> <li>不确定性采样</li> <li>人在回路</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>由放射科专家进行医学图像标注</li> <li>稀有方言的情感分析标记</li> <li>自动驾驶场景的选择</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/semi-supervised-learning-%E5%8D%8A%E7%9B%91%E7%9D%A3%E5%AD%A6%E4%B9%A0/">semi-supervised learning (半监督学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/query-strategies-%E6%9F%A5%E8%AF%A2%E7%AD%96%E7%95%A5/">query strategies (查询策略)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/data-annotation-%E6%95%B0%E6%8D%AE%E6%A0%87%E6%B3%A8/">data annotation (数据标注)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/label-efficiency-%E6%A0%87%E8%AE%B0%E6%95%88%E7%8E%87/">label efficiency (标记效率)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/zero_shot_learning/Sat, 18 Jul 2026 11:02:27 +0000https://terms-en.ai-term-hub.com/zh/terms/zero_shot_learning/<h2 id="definition">Definition</h2> <p>零样本学习使机器学习模型能够对训练数据集中未出现的类别实例进行分类。它不依赖每个可能类别的标记示例，而是利用类别之间的语义关系进行推理。</p> <h3 id="summary">Summary</h3> <p>零样本学习是一种技术，模型通过利用语义关联，对训练过程中从未见过的类别执行任务。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>泛化</li> <li>语义嵌入</li> <li>属性学习</li> <li>迁移学习</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>稀有动物物种分类</li> <li>新产品类别识别</li> <li>无需重新训练即可适应新兴威胁</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/few-shot-learning-%E5%B0%91%E6%A0%B7%E6%9C%AC%E5%AD%A6%E4%B9%A0/">Few-shot Learning (少样本学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transfer-learning-%E8%BF%81%E7%A7%BB%E5%AD%A6%E4%B9%A0/">Transfer Learning (迁移学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/semantic-embedding-%E8%AF%AD%E4%B9%89%E5%B5%8C%E5%85%A5/">Semantic Embedding (语义嵌入)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/generalization-%E6%B3%9B%E5%8C%96/">Generalization (泛化)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/qlora/Sat, 18 Jul 2026 11:01:29 +0000https://terms-en.ai-term-hub.com/zh/terms/qlora/<h2 id="definition">Definition</h2> <p>QLoRA 将低秩适应（LoRA）与 4 位量化相结合，显著减少了微调大规模模型所需的内存占用。通过将权重存储为 4 位格式并添加训练好的低秩适配器，该方法在保持性能的同时降低了资源需求。</p> <h3 id="summary">Summary</h3> <p>量化低秩适应，一种使用 4 位量化和低秩适配器高效微调大型语言模型的方法。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>低秩适应</li> <li>4 位量化</li> <li>内存效率</li> <li>微调</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>消费级 GPU 微调</li> <li>资源受限环境</li> <li>快速模型迭代</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><span style="color:#f92672">from</span> peft <span style="color:#f92672">import</span> LoraConfig, get_peft_model </span></span><span style="display:flex;"><span>config <span style="color:#f92672">=</span> LoraConfig(r<span style="color:#f92672">=</span><span style="color:#ae81ff">8</span>, lora_alpha<span style="color:#f92672">=</span><span style="color:#ae81ff">32</span>, target_modules<span style="color:#f92672">=</span>[<span style="color:#e6db74">&#34;q_proj&#34;</span>, <span style="color:#e6db74">&#34;v_proj&#34;</span>]) </span></span><span style="display:flex;"><span>model <span style="color:#f92672">=</span> get_peft_model(base_model, config) </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/lora/">LoRA</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%8F%82%E6%95%B0%E9%AB%98%E6%95%88%E5%BE%AE%E8%B0%83-peft/">参数高效微调 (PEFT)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E9%87%8F%E5%8C%96-quantization/">量化 (Quantization)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%8F%82%E6%95%B0%E9%AB%98%E6%95%88%E5%BE%AE%E8%B0%83-parameter-efficient-fine-tuning/">参数高效微调 (Parameter-Efficient Fine-Tuning)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/few_shot_learning/Sat, 18 Jul 2026 11:00:03 +0000https://terms-en.ai-term-hub.com/zh/terms/few_shot_learning/<h2 id="definition">Definition</h2> <p>少样本学习旨在使模型能够从寥寥几个示例中进行泛化，模仿人类的学习效率。它通常依赖于元学习策略，即模型先在一系列相关任务上进行预训练，从而获得快速适应新任务的能力，即使这些新任务只有少量标注数据可用。</p> <h3 id="summary">Summary</h3> <p>少样本学习是一种机器学习范式，模型仅从非常有限的标记训练数据中学习新概念。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>元学习</li> <li>数据效率</li> <li>泛化能力</li> <li>任务适配</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>罕见疾病识别</li> <li>新产品分类</li> <li>自定义物体检测</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/meta_learning-%E5%85%83%E5%AD%A6%E4%B9%A0/">meta_learning (元学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transfer_learning-%E8%BF%81%E7%A7%BB%E5%AD%A6%E4%B9%A0/">transfer_learning (迁移学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/data_efficiency-%E6%95%B0%E6%8D%AE%E6%95%88%E7%8E%87/">data_efficiency (数据效率)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/supervised_learning-%E7%9B%91%E7%9D%A3%E5%AD%A6%E4%B9%A0/">supervised_learning (监督学习)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/adapter/Sat, 18 Jul 2026 10:59:15 +0000https://terms-en.ai-term-hub.com/zh/terms/adapter/<h2 id="definition">Definition</h2> <p>适配器是一种参数高效的微调技术，主要用于大型语言模型和Transformer架构中。与计算成本高昂的更新所有模型权重不同，适配器仅训练少量新增参数，从而保留预训练知识并降低资源消耗。</p> <h3 id="summary">Summary</h3> <p>插入预训练模型中的轻量级模块，用于针对特定下游任务进行高效微调。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>参数高效微调</li> <li>迁移学习</li> <li>模块化架构</li> <li>灾难性遗忘</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>为客服聊天机器人微调大型语言模型</li> <li>适配视觉模型以进行医学图像分析</li> <li>高效部署多个领域特定的模型</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/lora-%E4%BD%8E%E7%A7%A9%E9%80%82%E5%BA%94/">LoRA (低秩适应)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/prompt-tuning-%E6%8F%90%E7%A4%BA%E5%BE%AE%E8%B0%83/">Prompt Tuning (提示微调)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/fine-tuning-%E5%BE%AE%E8%B0%83/">Fine-Tuning (微调)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transformer-transformer%E6%9E%B6%E6%9E%84/">Transformer (Transformer架构)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/training_free/Sat, 18 Jul 2026 10:59:01 +0000https://terms-en.ai-term-hub.com/zh/terms/training_free/<h2 id="definition">Definition</h2> <p>免训练方法是指在不通过反向传播更新底层权重的情况下修改模型行为或输出的技术。这些方法通常利用提示工程、特征提取或其他非参数化手段来实现适配。</p> <h3 id="summary">Summary</h3> <p>无需执行基于梯度的参数更新即可适应或增强模型的方法。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>提示工程</li> <li>无梯度下降</li> <li>参数高效</li> <li>零成本适配</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>通过提示进行少样本学习</li> <li>无需微调的领域自适应</li> <li>大语言模型应用的快速原型设计</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/prompting-%E6%8F%90%E7%A4%BA/">prompting (提示)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/fine_tuning-%E5%BE%AE%E8%B0%83/">fine_tuning (微调)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/in_context_learning-%E4%B8%8A%E4%B8%8B%E6%96%87%E5%AD%A6%E4%B9%A0/">in_context_learning (上下文学习)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/pre_trained/Sat, 18 Jul 2026 10:57:22 +0000https://terms-en.ai-term-hub.com/zh/terms/pre_trained/<h2 id="definition">Definition</h2> <p>预训练模型是一种基础人工智能模型，已在海量且多样化的数据集（如维基百科或ImageNet）上进行了广泛训练。这种初始训练使模型能够学习到广泛通用的特征表示。</p> <h3 id="summary">Summary</h3> <p>预训练模型是指已在大型数据集上训练过的神经网络，旨在学习通用特征，以便后续针对特定任务进行调整。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>迁移学习</li> <li>基础模型</li> <li>特征提取</li> <li>权重初始化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>使用大语言模型构建聊天机器人</li> <li>使用ResNet进行图像分类</li> <li>使用BERT进行情感分析</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/post_training-%E5%90%8E%E8%AE%AD%E7%BB%83/">post_training (后训练)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/foundation_model-%E5%9F%BA%E7%A1%80%E6%A8%A1%E5%9E%8B/">foundation_model (基础模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transfer_learning-%E8%BF%81%E7%A7%BB%E5%AD%A6%E4%B9%A0/">transfer_learning (迁移学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/weights-%E6%9D%83%E9%87%8D/">weights (权重)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/low_cost/Sat, 18 Jul 2026 10:56:47 +0000https://terms-en.ai-term-hub.com/zh/terms/low_cost/<h2 id="definition">Definition</h2> <p>低成本 AI 专注于效率，旨在降低与机器学习相关的入门壁垒和运营成本。这包括模型压缩、量化等技术&hellip;</p> <h3 id="summary">Summary</h3> <p>指在保持功能的同时最小化计算、财务或能源支出的 AI 解决方案。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>效率</li> <li>模型压缩</li> <li>边缘计算</li> <li>成本优化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>移动应用 AI 功能</li> <li>物联网传感器处理</li> <li>初创公司 MVP 开发</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/quantization-%E9%87%8F%E5%8C%96/">quantization (量化)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/edge_ai-%E8%BE%B9%E7%BC%98-ai/">edge_ai (边缘 AI)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/pruning-%E5%89%AA%E6%9E%9D/">pruning (剪枝)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/transfer_learning/Sat, 18 Jul 2026 10:55:40 +0000https://terms-en.ai-term-hub.com/zh/terms/transfer_learning/<h2 id="definition">Definition</h2> <p>迁移学习利用预训练模型来提高在新且相关任务上的性能并减少训练时间。开发人员无需从头开始训练，而是对现有权重进行微调，从而利用先前学到的知识。</p> <h3 id="summary">Summary</h3> <p>一种机器学习技术，将在一个任务中开发的模型作为第二个任务的模型的起点进行复用。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>预训练模型</li> <li>微调</li> <li>领域自适应</li> <li>特征提取</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>数据有限的图像分类</li> <li>针对小众主题的情感分析</li> <li>医疗诊断辅助</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><span style="color:#f92672">from</span> transformers <span style="color:#f92672">import</span> AutoModelForSequenceClassification </span></span><span style="display:flex;"><span>model <span style="color:#f92672">=</span> AutoModelForSequenceClassification<span style="color:#f92672">.</span>from_pretrained(<span style="color:#e6db74">&#39;bert-base-uncased&#39;</span>, num_labels<span style="color:#f92672">=</span><span style="color:#ae81ff">2</span>) </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/fine_tuning-%E5%BE%AE%E8%B0%83/">fine_tuning (微调)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/pre_training-%E9%A2%84%E8%AE%AD%E7%BB%83/">pre_training (预训练)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/domain_adaptation-%E9%A2%86%E5%9F%9F%E8%87%AA%E9%80%82%E5%BA%94/">domain_adaptation (领域自适应)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/few_shot_learning-%E5%B0%91%E6%A0%B7%E6%9C%AC%E5%AD%A6%E4%B9%A0/">few_shot_learning (少样本学习)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/mamba/Sat, 18 Jul 2026 10:53:03 +0000https://terms-en.ai-term-hub.com/zh/terms/mamba/<h2 id="definition">Definition</h2> <p>Mamba 通过引入硬件感知的选择性状态空间模型（SSM），代表了序列建模领域的重大进步。与随着序列长度呈二次方扩展的传统 Transformer 不同，Mamba 能够高效处理长序列。</p> <h3 id="summary">Summary</h3> <p>Mamba 是一种状态空间序列模型，它在保持 Transformer 在长上下文任务中性能的同时，提供了线性时间的推理速度。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>选择性状态空间模型</li> <li>线性复杂度</li> <li>硬件感知架构</li> <li>长上下文处理</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>长文档摘要</li> <li>基因组序列分析</li> <li>高频时间序列预测</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/transformer-%E5%8F%98%E6%8D%A2%E5%99%A8/">Transformer (变换器)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/rnn-%E5%BE%AA%E7%8E%AF%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C/">RNN (循环神经网络)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/ssm-%E7%8A%B6%E6%80%81%E7%A9%BA%E9%97%B4%E6%A8%A1%E5%9E%8B/">SSM (状态空间模型)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/attention-mechanism-%E6%B3%A8%E6%84%8F%E5%8A%9B%E6%9C%BA%E5%88%B6/">Attention Mechanism (注意力机制)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/lora/Sat, 18 Jul 2026 10:52:16 +0000https://terms-en.ai-term-hub.com/zh/terms/lora/<h2 id="definition">Definition</h2> <p>LoRA冻结预训练模型的权重，并在Transformer架构的每一层中插入可训练的分解矩阵。通过仅优化这些低秩矩阵，LoRA显著减少</p> <h3 id="summary">Summary</h3> <p>低秩自适应是一种参数高效的微调方法，它将可训练的秩分解矩阵注入现有模型权重中。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>参数高效微调</li> <li>秩分解</li> <li>权重冻结</li> <li>适配器模块</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>定制大语言模型</li> <li>领域特定适配</li> <li>资源受限的训练环境</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><span style="color:#f92672">from</span> peft <span style="color:#f92672">import</span> LoraConfig, get_peft_model </span></span><span style="display:flex;"><span>config <span style="color:#f92672">=</span> LoraConfig(r<span style="color:#f92672">=</span><span style="color:#ae81ff">8</span>, lora_alpha<span style="color:#f92672">=</span><span style="color:#ae81ff">32</span>) </span></span><span style="display:flex;"><span>model <span style="color:#f92672">=</span> get_peft_model(base_model, config) </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/peft-%E5%8F%82%E6%95%B0%E9%AB%98%E6%95%88%E5%BE%AE%E8%B0%83%E6%8A%80%E6%9C%AF/">PEFT (参数高效微调技术)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E5%BE%AE%E8%B0%83-fine-tuning/">微调 (Fine-Tuning)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/%E9%87%8F%E5%8C%96-quantization/">量化 (Quantization)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/direct/Sat, 18 Jul 2026 10:49:53 +0000https://terms-en.ai-term-hub.com/zh/terms/direct/<h2 id="definition">Definition</h2> <p>在AI语境中，“直接”通常描述绕过中间抽象层的架构或推理路径，例如强化学习中的直接策略优化或在端到端系统中直接映射输入到输出。</p> <h3 id="summary">Summary</h3> <p>指将输入直接映射到输出的方法或路径，无需经过中间复杂的变换或潜在表示。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>端到端</li> <li>降低延迟</li> <li>简洁性</li> <li>映射</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>实时控制系统</li> <li>简单回归任务</li> <li>边缘AI部署</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/end-to-end-learning-%E7%AB%AF%E5%88%B0%E7%AB%AF%E5%AD%A6%E4%B9%A0/">End-to-End Learning (端到端学习)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/regression-%E5%9B%9E%E5%BD%92/">Regression (回归)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/inference-%E6%8E%A8%E7%90%86/">Inference (推理)</a></li> </ul>https://terms-en.ai-term-hub.com/zh/terms/automated/Sat, 18 Jul 2026 10:49:18 +0000https://terms-en.ai-term-hub.com/zh/terms/automated/<h2 id="definition">Definition</h2> <p>AI中的自动化涉及使用算法和系统来执行传统上需要人力完成的任务。它通过执行预定义的规则或学习到的模式，专注于提高效率、一致性和速度。</p> <h3 id="summary">Summary</h3> <p>指由机器或软件以最少的人工干预执行的流程。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>人在回路</li> <li>效率</li> <li>基于规则的执行</li> <li>任务自动化</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>数据预处理流水线</li> <li>客户支持聊天机器人</li> <li>工业机器人装配</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/autonomous-%E8%87%AA%E4%B8%BB%E7%9A%84/">Autonomous (自主的)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/algorithm-%E7%AE%97%E6%B3%95/">Algorithm (算法)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/workflow-%E5%B7%A5%E4%BD%9C%E6%B5%81/">Workflow (工作流)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/orchestration-%E7%BC%96%E6%8E%92/">Orchestration (编排)</a></li> </ul>