Indexes: AI-Assisted Software Product / Materials¶
本页由
tools/gemini_build_material_indexes.py生成，用于给资料做标签、分类、按章节归档并排序。
Scoring¶
维度（0–10）：Relevance / Authority / Recency / Depth / Actionability。
加权总分：0.35·R + 0.20·A + 0.15·Re + 0.15·D + 0.15·Ac（范围 0–10）。
共 48 篇（已过滤 70 篇）。优先按章节分组；组内按总分降序。 低分清单：low-score.md
第 1 章：全流程方法论（01-method）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | [2308.08155] AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation | Agents & Tool Use | 多智能体, LLM框架, 对话编程, AutoGen, 人机协作 | note | link |
| 2 | 9.55 | [Overview | OpenAI Platform](sources/md/platform-openai-com-docs-00a8ca075d5c.md) | General References | OpenAI, GPT-5.2, 智能体, 模型优化, API文档 | note |
| 3 | 9.55 | Continuous Discovery Habits (the Book) is Finally Here! | Discovery & Product Strategy | 持续发现, 机会解决方案树, 产品发现, 成果导向, 假设测试 | note | link |
| 4 | 9.55 | Finding participants for user research - Service Manual - GOV.UK | Discovery & Product Strategy | 用户研究, 用户招募, 产品发现, 包容性设计, GOV.UK, 需求挖掘 | link | |
| 5 | 9.40 | LangGraph overview - Docs by LangChain | Agents & Tool Use | Agent 编排, 有状态工作流, LangGraph, LangChain, 人机协作 | note | link |
| 6 | 9.40 | How the discovery phase works - Service Manual - GOV.UK | Discovery & Product Strategy | 敏捷方法论, 发现阶段, 需求挖掘, GOV.UK, 服务设计 | link | |
| 7 | 9.25 | [2210.03629] ReAct: Synergizing Reasoning and Acting in Language Models | Agents & Tool Use | ReAct, 智能体, 思维链, 交互式决策, 推理与行动 | note | link |
| 8 | 9.10 | Learning about users and their needs - Service Manual - GOV.UK | Discovery & Product Strategy | 用户研究, 需求工程, GDS, 产品定义, 用户需求模板 | link | |
| 9 | 9.05 | Writing a user research plan - User Research Manual - Department for Education | Discovery & Product Strategy | 用户研究, 研究计划, 需求挖掘, 产品验证, 设计规范 | link | |
| 10 | 9.05 | Agile delivery - Service Manual - GOV.UK | General References | 敏捷开发, 项目管理, 政府服务设计, 交付流程, 用户故事 | link |
第 2 章：需求挖掘与机会判断（02-discovery）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | Continuous Discovery Habits (the Book) is Finally Here! | Discovery & Product Strategy | 持续发现, 机会解决方案树, 产品发现, 成果导向, 假设测试 | note | link |
| 2 | 9.55 | Finding participants for user research - Service Manual - GOV.UK | Discovery & Product Strategy | 用户研究, 用户招募, 产品发现, 包容性设计, GOV.UK, 需求挖掘 | link | |
| 3 | 9.40 | How the discovery phase works - Service Manual - GOV.UK | Discovery & Product Strategy | 敏捷方法论, 发现阶段, 需求挖掘, GOV.UK, 服务设计 | link | |
| 4 | 9.10 | Learning about users and their needs - Service Manual - GOV.UK | Discovery & Product Strategy | 用户研究, 需求工程, GDS, 产品定义, 用户需求模板 | link | |
| 5 | 9.05 | Writing a user research plan - User Research Manual - Department for Education | Discovery & Product Strategy | 用户研究, 研究计划, 需求挖掘, 产品验证, 设计规范 | link |
第 3 章：PRD 与工程合同（03-prd）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.70 | OpenAPI Specification v3.2.0 | PRD & Specs | API规范, OpenAPI, 契约驱动开发, 接口定义, Agent调用标准 | note | link |
| 2 | 9.55 | Overview | PRD & Specs | API Linting, OpenAPI, API 治理, 契约校验, JSON/YAML | note | link |
第 4 章：原型与信息架构（04-prototype）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.85 | Web Content Accessibility Guidelines (WCAG) 2.2 | UX / UI & Design Systems | 无障碍, WCAG, W3C标准, UX设计, 前端开发, 合规性 | note | link |
| 2 | 9.55 | Good from Afar, But Far from Good: AI Prototyping in Real Design Contexts - NN/G | UX / UI & Design Systems | AI 原型, UX 设计, UI 生成, 设计工作流, 可用性测试 | link | |
| 3 | 9.40 | PAIR Guidebook | UX / UI & Design Systems | 人机交互, UX 设计, AI 心理学, 设计规范 | note | link |
| 4 | 9.40 | Designing AI Products and Features: Study Guide - NN/G | UX / UI & Design Systems | AI产品设计, UX/UI, 价值主张, 提示词辅助, 生成式UI, NN/g | link | |
| 5 | 9.25 | Guidelines for Human-AI Interaction - Microsoft Research | UX / UI & Design Systems | 人机交互, AI设计准则, 用户体验, HAX Toolkit, 负责任AI | note | link |
| 6 | 9.20 | Design Tokens Format Module 2025.10 | UX / UI & Design Systems | 设计令牌, 设计系统, W3C规范草案, JSON交换格式, UI工程化 | note | link |
| 7 | 9.20 | axe-core | UX / UI & Design Systems | 可访问性, 自动化测试, WCAG, 前端开发, UI | note | link |
第 5 章：产品验证与打磨（05-validation）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.85 | Web Content Accessibility Guidelines (WCAG) 2.2 | UX / UI & Design Systems | 无障碍, WCAG, W3C标准, UX设计, 前端开发, 合规性 | note | link |
| 2 | 9.55 | Continuous Discovery Habits (the Book) is Finally Here! | Discovery & Product Strategy | 持续发现, 机会解决方案树, 产品发现, 成果导向, 假设测试 | note | link |
| 3 | 9.55 | Finding participants for user research - Service Manual - GOV.UK | Discovery & Product Strategy | 用户研究, 用户招募, 产品发现, 包容性设计, GOV.UK, 需求挖掘 | link | |
| 4 | 9.55 | Good from Afar, But Far from Good: AI Prototyping in Real Design Contexts - NN/G | UX / UI & Design Systems | AI 原型, UX 设计, UI 生成, 设计工作流, 可用性测试 | link | |
| 5 | 9.40 | PAIR Guidebook | UX / UI & Design Systems | 人机交互, UX 设计, AI 心理学, 设计规范 | note | link |
| 6 | 9.40 | How the discovery phase works - Service Manual - GOV.UK | Discovery & Product Strategy | 敏捷方法论, 发现阶段, 需求挖掘, GOV.UK, 服务设计 | link | |
| 7 | 9.40 | Designing AI Products and Features: Study Guide - NN/G | UX / UI & Design Systems | AI产品设计, UX/UI, 价值主张, 提示词辅助, 生成式UI, NN/g | link | |
| 8 | 9.25 | Guidelines for Human-AI Interaction - Microsoft Research | UX / UI & Design Systems | 人机交互, AI设计准则, 用户体验, HAX Toolkit, 负责任AI | note | link |
| 9 | 9.20 | Design Tokens Format Module 2025.10 | UX / UI & Design Systems | 设计令牌, 设计系统, W3C规范草案, JSON交换格式, UI工程化 | note | link |
| 10 | 9.20 | axe-core | UX / UI & Design Systems | 可访问性, 自动化测试, WCAG, 前端开发, UI | note | link |
| 11 | 9.10 | Learning about users and their needs - Service Manual - GOV.UK | Discovery & Product Strategy | 用户研究, 需求工程, GDS, 产品定义, 用户需求模板 | link | |
| 12 | 9.05 | Writing a user research plan - User Research Manual - Department for Education | Discovery & Product Strategy | 用户研究, 研究计划, 需求挖掘, 产品验证, 设计规范 | link |
第 6 章：UI 设计与资产化（06-ui）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.85 | Web Content Accessibility Guidelines (WCAG) 2.2 | UX / UI & Design Systems | 无障碍, WCAG, W3C标准, UX设计, 前端开发, 合规性 | note | link |
| 2 | 9.55 | Good from Afar, But Far from Good: AI Prototyping in Real Design Contexts - NN/G | UX / UI & Design Systems | AI 原型, UX 设计, UI 生成, 设计工作流, 可用性测试 | link | |
| 3 | 9.40 | PAIR Guidebook | UX / UI & Design Systems | 人机交互, UX 设计, AI 心理学, 设计规范 | note | link |
| 4 | 9.40 | Designing AI Products and Features: Study Guide - NN/G | UX / UI & Design Systems | AI产品设计, UX/UI, 价值主张, 提示词辅助, 生成式UI, NN/g | link | |
| 5 | 9.25 | Guidelines for Human-AI Interaction - Microsoft Research | UX / UI & Design Systems | 人机交互, AI设计准则, 用户体验, HAX Toolkit, 负责任AI | note | link |
| 6 | 9.20 | Design Tokens Format Module 2025.10 | UX / UI & Design Systems | 设计令牌, 设计系统, W3C规范草案, JSON交换格式, UI工程化 | note | link |
| 7 | 9.20 | axe-core | UX / UI & Design Systems | 可访问性, 自动化测试, WCAG, 前端开发, UI | note | link |
第 7 章：工程化与编码（07-engineering）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.70 | OpenAPI Specification v3.2.0 | PRD & Specs | API规范, OpenAPI, 契约驱动开发, 接口定义, Agent调用标准 | note | link |
| 2 | 9.55 | Final: OpenID Connect Core 1.0 incorporating errata set 2 | User / Auth / Audit | OIDC, OAuth 2.0, 身份认证, ID Token, 用户声明 | note | link |
| 3 | 9.55 | [2308.08155] AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation | Agents & Tool Use | 多智能体, LLM框架, 对话编程, AutoGen, 人机协作 | note | link |
| 4 | 9.55 | [2309.06180] Efficient Memory Management for Large Language Model Serving with PagedAttention | Inference & Optimization | vLLM, PagedAttention, KV Cache, 推理优化, 内存管理 | note | link |
| 5 | 9.55 | Overview | PRD & Specs | API Linting, OpenAPI, API 治理, 契约校验, JSON/YAML | note | link |
| 6 | 9.55 | Google SRE - Defining slo: service level objective meaning | Deployment & Operations | SRE, SLO, SLI, 系统可靠性, 错误预算, 可观测性 | note | link |
| 7 | 9.50 | [Grafana OSS and Enterprise | Grafana documentation](sources/md/grafana-com-docs-grafana-latest-02ef4859d21b.md) | Deployment & Operations | 可观测性, 监控, 可视化, 运维, 指标分析 | note |
| 8 | 9.40 | LangGraph overview - Docs by LangChain | Agents & Tool Use | Agent 编排, 有状态工作流, LangGraph, LangChain, 人机协作 | note | link |
| 9 | 9.40 | github-com-nvidia-tensorrt-llm | Inference & Optimization | LLM推理, GPU加速, 量化, NVIDIA, 深度学习框架 | link | |
| 10 | 9.40 | [2210.17323] GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers | Inference & Optimization | 模型量化, GPTQ, LLM部署, 推理加速, 后训练量化 | note | link |
| 11 | 9.40 | KServe | Inference & Optimization | AI推理, Kubernetes, 模型服务, LLM运维, CNCF | note | link |
| 12 | 9.40 | [2306.05685] Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena | Evaluation | LLM-as-a-Judge, MT-bench, Chatbot Arena, 模型评测, 人类偏好对齐 | note | link |
| 13 | 9.35 | [OWASP Application Security Verification Standard (ASVS) | OWASP Foundation](sources/md/owasp-org-www-project-application-security-verification-standard-50b4bc9ed328.md) | Engineering Workflow | Web安全, 安全标准, OWASP, 安全开发, 渗透测试 | note |
| 14 | 9.35 | [Documentation | OpenTelemetry](sources/md/opentelemetry-io-docs-8de8f872c85e.md) | Deployment & Operations | 可观测性, 云原生, OpenTelemetry, 分布式追踪, GenAI监控 | note |
| 15 | 9.25 | RFC 7519: JSON Web Token (JWT) | User / Auth / Audit | JWT, RFC, 身份认证, API安全, JSON | note | link |
| 16 | 9.25 | [2210.03629] ReAct: Synergizing Reasoning and Acting in Language Models | Agents & Tool Use | ReAct, 智能体, 思维链, 交互式决策, 推理与行动 | note | link |
| 17 | 9.25 | Text Generation Inference | Inference & Optimization | LLM 推理, 模型部署, Hugging Face, 量化优化, 高并发 | note | link |
| 18 | 9.20 | [Lighthouse | Chrome for Developers](sources/md/developer-chrome-com-docs-lighthouse-1b45ffb23796.md) | Evaluation | 网页审计, 前端性能, 可访问性, SEO, 自动化测试 | note |
| 19 | 9.20 | [Fast and reliable end-to-end testing for modern web apps | Playwright](sources/md/playwright-dev-8960df5ea200.md) | Engineering Workflow | 端到端测试, 自动化测试, Playwright, 软件工程, 前端开发 | note |
| 20 | 9.20 | [Test runner | Storybook docs](sources/md/storybook-js-org-docs-writing-tests-integrations-test-runner-28169fa9753f.md) | Engineering Workflow | Storybook, 自动化测试, Playwright, 组件驱动开发, CI/CD | note |
| 21 | 9.20 | [2309.15217] Ragas: Automated Evaluation of Retrieval Augmented Generation | Evaluation | RAG, 自动化评估, LLM, Ragas, 检索增强生成 | note | link |
| 22 | 9.20 | RFC 9457: Problem Details for HTTP APIs | Backend & Reliability | HTTP API, 错误处理, 标准规范, RFC 9457, JSON | note | link |
| 23 | 9.10 | RFC 6749: The OAuth 2.0 Authorization Framework | User / Auth / Audit | OAuth 2.0, 授权框架, 身份认证, API 安全, RFC 标准 | note | link |
| 24 | 9.10 | [2005.11401] Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks | RAG | RAG, 检索增强, DPR, BART, 知识图谱, 深度学习, 事实准确性 | note | link |
| 25 | 9.05 | Set up a service team at each phase - Service Manual - GOV.UK | Engineering Workflow | 敏捷开发, 多学科团队, 团队角色, GDS, 产品生命周期, 交付管理 | link |
第 8 章：前端实现（08-frontend）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.85 | Web Content Accessibility Guidelines (WCAG) 2.2 | UX / UI & Design Systems | 无障碍, WCAG, W3C标准, UX设计, 前端开发, 合规性 | note | link |
| 2 | 9.55 | Good from Afar, But Far from Good: AI Prototyping in Real Design Contexts - NN/G | UX / UI & Design Systems | AI 原型, UX 设计, UI 生成, 设计工作流, 可用性测试 | link | |
| 3 | 9.40 | PAIR Guidebook | UX / UI & Design Systems | 人机交互, UX 设计, AI 心理学, 设计规范 | note | link |
| 4 | 9.40 | Designing AI Products and Features: Study Guide - NN/G | UX / UI & Design Systems | AI产品设计, UX/UI, 价值主张, 提示词辅助, 生成式UI, NN/g | link | |
| 5 | 9.35 | [OWASP Application Security Verification Standard (ASVS) | OWASP Foundation](sources/md/owasp-org-www-project-application-security-verification-standard-50b4bc9ed328.md) | Engineering Workflow | Web安全, 安全标准, OWASP, 安全开发, 渗透测试 | note |
| 6 | 9.25 | Guidelines for Human-AI Interaction - Microsoft Research | UX / UI & Design Systems | 人机交互, AI设计准则, 用户体验, HAX Toolkit, 负责任AI | note | link |
| 7 | 9.20 | Design Tokens Format Module 2025.10 | UX / UI & Design Systems | 设计令牌, 设计系统, W3C规范草案, JSON交换格式, UI工程化 | note | link |
| 8 | 9.20 | axe-core | UX / UI & Design Systems | 可访问性, 自动化测试, WCAG, 前端开发, UI | note | link |
| 9 | 9.20 | [Fast and reliable end-to-end testing for modern web apps | Playwright](sources/md/playwright-dev-8960df5ea200.md) | Engineering Workflow | 端到端测试, 自动化测试, Playwright, 软件工程, 前端开发 | note |
| 10 | 9.20 | [Test runner | Storybook docs](sources/md/storybook-js-org-docs-writing-tests-integrations-test-runner-28169fa9753f.md) | Engineering Workflow | Storybook, 自动化测试, Playwright, 组件驱动开发, CI/CD | note |
| 11 | 9.05 | Set up a service team at each phase - Service Manual - GOV.UK | Engineering Workflow | 敏捷开发, 多学科团队, 团队角色, GDS, 产品生命周期, 交付管理 | link |
第 9 章：后端架构（09-backend）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.70 | OpenAPI Specification v3.2.0 | PRD & Specs | API规范, OpenAPI, 契约驱动开发, 接口定义, Agent调用标准 | note | link |
| 2 | 9.55 | Final: OpenID Connect Core 1.0 incorporating errata set 2 | User / Auth / Audit | OIDC, OAuth 2.0, 身份认证, ID Token, 用户声明 | note | link |
| 3 | 9.55 | Overview | PRD & Specs | API Linting, OpenAPI, API 治理, 契约校验, JSON/YAML | note | link |
| 4 | 9.35 | [OWASP Application Security Verification Standard (ASVS) | OWASP Foundation](sources/md/owasp-org-www-project-application-security-verification-standard-50b4bc9ed328.md) | Engineering Workflow | Web安全, 安全标准, OWASP, 安全开发, 渗透测试 | note |
| 5 | 9.25 | RFC 7519: JSON Web Token (JWT) | User / Auth / Audit | JWT, RFC, 身份认证, API安全, JSON | note | link |
| 6 | 9.25 | [従量課金 | Stripe のドキュメント](sources/md/docs-stripe-com-billing-subscriptions-usage-based-06655737082c.md) | Billing & Pricing | Stripe, 按量计费, SaaS, 支付系统, Token计费 | note |
| 7 | 9.20 | PCI Security Standards Council – Protect Payment Data with Industry-driven Security Standards, Training, and Programs | Billing & Pricing | PCI DSS, 支付安全, 合规性, 数据保护, 金融标准 | note | link |
| 8 | 9.20 | [Fast and reliable end-to-end testing for modern web apps | Playwright](sources/md/playwright-dev-8960df5ea200.md) | Engineering Workflow | 端到端测试, 自动化测试, Playwright, 软件工程, 前端开发 | note |
| 9 | 9.20 | [Test runner | Storybook docs](sources/md/storybook-js-org-docs-writing-tests-integrations-test-runner-28169fa9753f.md) | Engineering Workflow | Storybook, 自动化测试, Playwright, 组件驱动开发, CI/CD | note |
| 10 | 9.20 | RFC 9457: Problem Details for HTTP APIs | Backend & Reliability | HTTP API, 错误处理, 标准规范, RFC 9457, JSON | note | link |
| 11 | 9.10 | RFC 6749: The OAuth 2.0 Authorization Framework | User / Auth / Audit | OAuth 2.0, 授权框架, 身份认证, API 安全, RFC 标准 | note | link |
| 12 | 9.05 | Set up a service team at each phase - Service Manual - GOV.UK | Engineering Workflow | 敏捷开发, 多学科团队, 团队角色, GDS, 产品生命周期, 交付管理 | link |
第 10 章：RAG & Agent（10-intelligence）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.70 | OpenAPI Specification v3.2.0 | PRD & Specs | API规范, OpenAPI, 契约驱动开发, 接口定义, Agent调用标准 | note | link |
| 2 | 9.55 | [2308.08155] AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation | Agents & Tool Use | 多智能体, LLM框架, 对话编程, AutoGen, 人机协作 | note | link |
| 3 | 9.55 | Overview | PRD & Specs | API Linting, OpenAPI, API 治理, 契约校验, JSON/YAML | note | link |
| 4 | 9.40 | LangGraph overview - Docs by LangChain | Agents & Tool Use | Agent 编排, 有状态工作流, LangGraph, LangChain, 人机协作 | note | link |
| 5 | 9.40 | [2306.05685] Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena | Evaluation | LLM-as-a-Judge, MT-bench, Chatbot Arena, 模型评测, 人类偏好对齐 | note | link |
| 6 | 9.25 | [2210.03629] ReAct: Synergizing Reasoning and Acting in Language Models | Agents & Tool Use | ReAct, 智能体, 思维链, 交互式决策, 推理与行动 | note | link |
| 7 | 9.25 | [1803.09010] Datasheets for Datasets | Data | 数据集文档, 数据治理, 透明度, 机器学习偏见, 数据管理 | note | link |
| 8 | 9.20 | [Lighthouse | Chrome for Developers](sources/md/developer-chrome-com-docs-lighthouse-1b45ffb23796.md) | Evaluation | 网页审计, 前端性能, 可访问性, SEO, 自动化测试 | note |
| 9 | 9.20 | [2309.15217] Ragas: Automated Evaluation of Retrieval Augmented Generation | Evaluation | RAG, 自动化评估, LLM, Ragas, 检索增强生成 | note | link |
| 10 | 9.10 | [2005.11401] Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks | RAG | RAG, 检索增强, DPR, BART, 知识图谱, 深度学习, 事实准确性 | note | link |
第 11 章：用户与权限（11-user）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | Final: OpenID Connect Core 1.0 incorporating errata set 2 | User / Auth / Audit | OIDC, OAuth 2.0, 身份认证, ID Token, 用户声明 | note | link |
| 2 | 9.55 | [AI Risk Management Framework | NIST](sources/md/www-nist-gov-itl-ai-risk-management-framework-54dbc1541342.md) | Governance & Security | AI治理, 风险管理, NIST标准, 可信AI, 合规性 | note |
| 3 | 9.25 | RFC 7519: JSON Web Token (JWT) | User / Auth / Audit | JWT, RFC, 身份认证, API安全, JSON | note | link |
| 4 | 9.10 | RFC 6749: The OAuth 2.0 Authorization Framework | User / Auth / Audit | OAuth 2.0, 授权框架, 身份认证, API 安全, RFC 标准 | note | link |
第 12 章：付费与风控（12-billing）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | [AI Risk Management Framework | NIST](sources/md/www-nist-gov-itl-ai-risk-management-framework-54dbc1541342.md) | Governance & Security | AI治理, 风险管理, NIST标准, 可信AI, 合规性 | note |
| 2 | 9.25 | [従量課金 | Stripe のドキュメント](sources/md/docs-stripe-com-billing-subscriptions-usage-based-06655737082c.md) | Billing & Pricing | Stripe, 按量计费, SaaS, 支付系统, Token计费 | note |
| 3 | 9.20 | PCI Security Standards Council – Protect Payment Data with Industry-driven Security Standards, Training, and Programs | Billing & Pricing | PCI DSS, 支付安全, 合规性, 数据保护, 金融标准 | note | link |
第 13 章：数据收集与清洗（13-data）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.70 | [2305.18290] Direct Preference Optimization: Your Language Model is Secretly a Reward Model | Training & Alignment | DPO, RLHF, 模型对齐, LLM, 微调 | note | link |
| 2 | 9.55 | [2106.09685] LoRA: Low-Rank Adaptation of Large Language Models | Training & Alignment | LoRA, PEFT, 大模型微调, 参数高效, 模型压缩 | note | link |
| 3 | 9.55 | [2305.14314] QLoRA: Efficient Finetuning of Quantized LLMs | Training & Alignment | QLoRA, 低比特微调, 显存优化, 量化技术, Guanaco | note | link |
| 4 | 9.55 | [AI Risk Management Framework | NIST](sources/md/www-nist-gov-itl-ai-risk-management-framework-54dbc1541342.md) | Governance & Security | AI治理, 风险管理, NIST标准, 可信AI, 合规性 | note |
| 5 | 9.40 | [2212.10560] Self-Instruct: Aligning Language Models with Self-Generated Instructions | Training & Alignment | Self-Instruct, 合成数据, 指令微调, 模型对齐, 数据管道 | note | link |
| 6 | 9.25 | [1803.09010] Datasheets for Datasets | Data | 数据集文档, 数据治理, 透明度, 机器学习偏见, 数据管理 | note | link |
| 7 | 9.10 | [2005.11401] Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks | RAG | RAG, 检索增强, DPR, BART, 知识图谱, 深度学习, 事实准确性 | note | link |
| 8 | 9.10 | [2203.02155] Training language models to follow instructions with human feedback | Training & Alignment | RLHF, InstructGPT, 模型对齐, PPO, 指令遵循 | note | link |
第 14 章：预训练（14-pretrain）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.70 | [2305.18290] Direct Preference Optimization: Your Language Model is Secretly a Reward Model | Training & Alignment | DPO, RLHF, 模型对齐, LLM, 微调 | note | link |
| 2 | 9.55 | [2106.09685] LoRA: Low-Rank Adaptation of Large Language Models | Training & Alignment | LoRA, PEFT, 大模型微调, 参数高效, 模型压缩 | note | link |
| 3 | 9.55 | [2305.14314] QLoRA: Efficient Finetuning of Quantized LLMs | Training & Alignment | QLoRA, 低比特微调, 显存优化, 量化技术, Guanaco | note | link |
| 4 | 9.40 | [2212.10560] Self-Instruct: Aligning Language Models with Self-Generated Instructions | Training & Alignment | Self-Instruct, 合成数据, 指令微调, 模型对齐, 数据管道 | note | link |
| 5 | 9.10 | [2203.02155] Training language models to follow instructions with human feedback | Training & Alignment | RLHF, InstructGPT, 模型对齐, PPO, 指令遵循 | note | link |
第 15 章：后训练与对齐（15-posttrain）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.70 | [2305.18290] Direct Preference Optimization: Your Language Model is Secretly a Reward Model | Training & Alignment | DPO, RLHF, 模型对齐, LLM, 微调 | note | link |
| 2 | 9.55 | [2106.09685] LoRA: Low-Rank Adaptation of Large Language Models | Training & Alignment | LoRA, PEFT, 大模型微调, 参数高效, 模型压缩 | note | link |
| 3 | 9.55 | [2305.14314] QLoRA: Efficient Finetuning of Quantized LLMs | Training & Alignment | QLoRA, 低比特微调, 显存优化, 量化技术, Guanaco | note | link |
| 4 | 9.40 | [2212.10560] Self-Instruct: Aligning Language Models with Self-Generated Instructions | Training & Alignment | Self-Instruct, 合成数据, 指令微调, 模型对齐, 数据管道 | note | link |
| 5 | 9.10 | [2203.02155] Training language models to follow instructions with human feedback | Training & Alignment | RLHF, InstructGPT, 模型对齐, PPO, 指令遵循 | note | link |
第 16 章：推理优化（16-inference）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | [2309.06180] Efficient Memory Management for Large Language Model Serving with PagedAttention | Inference & Optimization | vLLM, PagedAttention, KV Cache, 推理优化, 内存管理 | note | link |
| 2 | 9.40 | github-com-nvidia-tensorrt-llm | Inference & Optimization | LLM推理, GPU加速, 量化, NVIDIA, 深度学习框架 | link | |
| 3 | 9.40 | [2210.17323] GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers | Inference & Optimization | 模型量化, GPTQ, LLM部署, 推理加速, 后训练量化 | note | link |
| 4 | 9.40 | KServe | Inference & Optimization | AI推理, Kubernetes, 模型服务, LLM运维, CNCF | note | link |
| 5 | 9.25 | Text Generation Inference | Inference & Optimization | LLM 推理, 模型部署, Hugging Face, 量化优化, 高并发 | note | link |
第 17 章：部署与运维（17-deployment）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | [2309.06180] Efficient Memory Management for Large Language Model Serving with PagedAttention | Inference & Optimization | vLLM, PagedAttention, KV Cache, 推理优化, 内存管理 | note | link |
| 2 | 9.55 | Google SRE - Defining slo: service level objective meaning | Deployment & Operations | SRE, SLO, SLI, 系统可靠性, 错误预算, 可观测性 | note | link |
| 3 | 9.50 | [Grafana OSS and Enterprise | Grafana documentation](sources/md/grafana-com-docs-grafana-latest-02ef4859d21b.md) | Deployment & Operations | 可观测性, 监控, 可视化, 运维, 指标分析 | note |
| 4 | 9.40 | github-com-nvidia-tensorrt-llm | Inference & Optimization | LLM推理, GPU加速, 量化, NVIDIA, 深度学习框架 | link | |
| 5 | 9.40 | [2210.17323] GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers | Inference & Optimization | 模型量化, GPTQ, LLM部署, 推理加速, 后训练量化 | note | link |
| 6 | 9.40 | KServe | Inference & Optimization | AI推理, Kubernetes, 模型服务, LLM运维, CNCF | note | link |
| 7 | 9.40 | [2306.05685] Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena | Evaluation | LLM-as-a-Judge, MT-bench, Chatbot Arena, 模型评测, 人类偏好对齐 | note | link |
| 8 | 9.35 | [OWASP Application Security Verification Standard (ASVS) | OWASP Foundation](sources/md/owasp-org-www-project-application-security-verification-standard-50b4bc9ed328.md) | Engineering Workflow | Web安全, 安全标准, OWASP, 安全开发, 渗透测试 | note |
| 9 | 9.35 | [Documentation | OpenTelemetry](sources/md/opentelemetry-io-docs-8de8f872c85e.md) | Deployment & Operations | 可观测性, 云原生, OpenTelemetry, 分布式追踪, GenAI监控 | note |
| 10 | 9.25 | Text Generation Inference | Inference & Optimization | LLM 推理, 模型部署, Hugging Face, 量化优化, 高并发 | note | link |
| 11 | 9.20 | [Lighthouse | Chrome for Developers](sources/md/developer-chrome-com-docs-lighthouse-1b45ffb23796.md) | Evaluation | 网页审计, 前端性能, 可访问性, SEO, 自动化测试 | note |
| 12 | 9.20 | [Fast and reliable end-to-end testing for modern web apps | Playwright](sources/md/playwright-dev-8960df5ea200.md) | Engineering Workflow | 端到端测试, 自动化测试, Playwright, 软件工程, 前端开发 | note |
| 13 | 9.20 | [Test runner | Storybook docs](sources/md/storybook-js-org-docs-writing-tests-integrations-test-runner-28169fa9753f.md) | Engineering Workflow | Storybook, 自动化测试, Playwright, 组件驱动开发, CI/CD | note |
| 14 | 9.20 | [2309.15217] Ragas: Automated Evaluation of Retrieval Augmented Generation | Evaluation | RAG, 自动化评估, LLM, Ragas, 检索增强生成 | note | link |
| 15 | 9.20 | RFC 9457: Problem Details for HTTP APIs | Backend & Reliability | HTTP API, 错误处理, 标准规范, RFC 9457, JSON | note | link |
| 16 | 9.05 | Set up a service team at each phase - Service Manual - GOV.UK | Engineering Workflow | 敏捷开发, 多学科团队, 团队角色, GDS, 产品生命周期, 交付管理 | link |
第 18 章：评测体系（18-evaluation）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.70 | [2305.18290] Direct Preference Optimization: Your Language Model is Secretly a Reward Model | Training & Alignment | DPO, RLHF, 模型对齐, LLM, 微调 | note | link |
| 2 | 9.55 | [2308.08155] AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation | Agents & Tool Use | 多智能体, LLM框架, 对话编程, AutoGen, 人机协作 | note | link |
| 3 | 9.55 | [2106.09685] LoRA: Low-Rank Adaptation of Large Language Models | Training & Alignment | LoRA, PEFT, 大模型微调, 参数高效, 模型压缩 | note | link |
| 4 | 9.55 | [2305.14314] QLoRA: Efficient Finetuning of Quantized LLMs | Training & Alignment | QLoRA, 低比特微调, 显存优化, 量化技术, Guanaco | note | link |
| 5 | 9.55 | [2309.06180] Efficient Memory Management for Large Language Model Serving with PagedAttention | Inference & Optimization | vLLM, PagedAttention, KV Cache, 推理优化, 内存管理 | note | link |
| 6 | 9.55 | Google SRE - Defining slo: service level objective meaning | Deployment & Operations | SRE, SLO, SLI, 系统可靠性, 错误预算, 可观测性 | note | link |
| 7 | 9.50 | [Grafana OSS and Enterprise | Grafana documentation](sources/md/grafana-com-docs-grafana-latest-02ef4859d21b.md) | Deployment & Operations | 可观测性, 监控, 可视化, 运维, 指标分析 | note |
| 8 | 9.40 | LangGraph overview - Docs by LangChain | Agents & Tool Use | Agent 编排, 有状态工作流, LangGraph, LangChain, 人机协作 | note | link |
| 9 | 9.40 | [2212.10560] Self-Instruct: Aligning Language Models with Self-Generated Instructions | Training & Alignment | Self-Instruct, 合成数据, 指令微调, 模型对齐, 数据管道 | note | link |
| 10 | 9.40 | github-com-nvidia-tensorrt-llm | Inference & Optimization | LLM推理, GPU加速, 量化, NVIDIA, 深度学习框架 | link | |
| 11 | 9.40 | [2210.17323] GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers | Inference & Optimization | 模型量化, GPTQ, LLM部署, 推理加速, 后训练量化 | note | link |
| 12 | 9.40 | KServe | Inference & Optimization | AI推理, Kubernetes, 模型服务, LLM运维, CNCF | note | link |
| 13 | 9.40 | [2306.05685] Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena | Evaluation | LLM-as-a-Judge, MT-bench, Chatbot Arena, 模型评测, 人类偏好对齐 | note | link |
| 14 | 9.35 | [Documentation | OpenTelemetry](sources/md/opentelemetry-io-docs-8de8f872c85e.md) | Deployment & Operations | 可观测性, 云原生, OpenTelemetry, 分布式追踪, GenAI监控 | note |
| 15 | 9.25 | [2210.03629] ReAct: Synergizing Reasoning and Acting in Language Models | Agents & Tool Use | ReAct, 智能体, 思维链, 交互式决策, 推理与行动 | note | link |
| 16 | 9.25 | [1803.09010] Datasheets for Datasets | Data | 数据集文档, 数据治理, 透明度, 机器学习偏见, 数据管理 | note | link |
| 17 | 9.25 | Text Generation Inference | Inference & Optimization | LLM 推理, 模型部署, Hugging Face, 量化优化, 高并发 | note | link |
| 18 | 9.20 | [Lighthouse | Chrome for Developers](sources/md/developer-chrome-com-docs-lighthouse-1b45ffb23796.md) | Evaluation | 网页审计, 前端性能, 可访问性, SEO, 自动化测试 | note |
| 19 | 9.20 | [2309.15217] Ragas: Automated Evaluation of Retrieval Augmented Generation | Evaluation | RAG, 自动化评估, LLM, Ragas, 检索增强生成 | note | link |
| 20 | 9.20 | RFC 9457: Problem Details for HTTP APIs | Backend & Reliability | HTTP API, 错误处理, 标准规范, RFC 9457, JSON | note | link |
| 21 | 9.10 | [2005.11401] Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks | RAG | RAG, 检索增强, DPR, BART, 知识图谱, 深度学习, 事实准确性 | note | link |
| 22 | 9.10 | [2203.02155] Training language models to follow instructions with human feedback | Training & Alignment | RLHF, InstructGPT, 模型对齐, PPO, 指令遵循 | note | link |
第 19 章：迭代与增长（19-iteration）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | Continuous Discovery Habits (the Book) is Finally Here! | Discovery & Product Strategy | 持续发现, 机会解决方案树, 产品发现, 成果导向, 假设测试 | note | link |
| 2 | 9.55 | Finding participants for user research - Service Manual - GOV.UK | Discovery & Product Strategy | 用户研究, 用户招募, 产品发现, 包容性设计, GOV.UK, 需求挖掘 | link | |
| 3 | 9.40 | How the discovery phase works - Service Manual - GOV.UK | Discovery & Product Strategy | 敏捷方法论, 发现阶段, 需求挖掘, GOV.UK, 服务设计 | link | |
| 4 | 9.25 | [従量課金 | Stripe のドキュメント](sources/md/docs-stripe-com-billing-subscriptions-usage-based-06655737082c.md) | Billing & Pricing | Stripe, 按量计费, SaaS, 支付系统, Token计费 | note |
| 5 | 9.20 | PCI Security Standards Council – Protect Payment Data with Industry-driven Security Standards, Training, and Programs | Billing & Pricing | PCI DSS, 支付安全, 合规性, 数据保护, 金融标准 | note | link |
| 6 | 9.10 | Learning about users and their needs - Service Manual - GOV.UK | Discovery & Product Strategy | 用户研究, 需求工程, GDS, 产品定义, 用户需求模板 | link | |
| 7 | 9.05 | Writing a user research plan - User Research Manual - Department for Education | Discovery & Product Strategy | 用户研究, 研究计划, 需求挖掘, 产品验证, 设计规范 | link |
第 20 章：合规与伦理（20-governance）¶
| # | Score | Title | Category | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | Final: OpenID Connect Core 1.0 incorporating errata set 2 | User / Auth / Audit | OIDC, OAuth 2.0, 身份认证, ID Token, 用户声明 | note | link |
| 2 | 9.55 | Google SRE - Defining slo: service level objective meaning | Deployment & Operations | SRE, SLO, SLI, 系统可靠性, 错误预算, 可观测性 | note | link |
| 3 | 9.55 | [AI Risk Management Framework | NIST](sources/md/www-nist-gov-itl-ai-risk-management-framework-54dbc1541342.md) | Governance & Security | AI治理, 风险管理, NIST标准, 可信AI, 合规性 | note |
| 4 | 9.50 | [Grafana OSS and Enterprise | Grafana documentation](sources/md/grafana-com-docs-grafana-latest-02ef4859d21b.md) | Deployment & Operations | 可观测性, 监控, 可视化, 运维, 指标分析 | note |
| 5 | 9.35 | [Documentation | OpenTelemetry](sources/md/opentelemetry-io-docs-8de8f872c85e.md) | Deployment & Operations | 可观测性, 云原生, OpenTelemetry, 分布式追踪, GenAI监控 | note |
| 6 | 9.25 | RFC 7519: JSON Web Token (JWT) | User / Auth / Audit | JWT, RFC, 身份认证, API安全, JSON | note | link |
| 7 | 9.25 | [1803.09010] Datasheets for Datasets | Data | 数据集文档, 数据治理, 透明度, 机器学习偏见, 数据管理 | note | link |
| 8 | 9.25 | [従量課金 | Stripe のドキュメント](sources/md/docs-stripe-com-billing-subscriptions-usage-based-06655737082c.md) | Billing & Pricing | Stripe, 按量计费, SaaS, 支付系统, Token计费 | note |
| 9 | 9.20 | PCI Security Standards Council – Protect Payment Data with Industry-driven Security Standards, Training, and Programs | Billing & Pricing | PCI DSS, 支付安全, 合规性, 数据保护, 金融标准 | note | link |
| 10 | 9.10 | RFC 6749: The OAuth 2.0 Authorization Framework | User / Auth / Audit | OAuth 2.0, 授权框架, 身份认证, API 安全, RFC 标准 | note | link |
按领域分类（补充视角）¶
Discovery & Product Strategy（product_discovery）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | Continuous Discovery Habits (the Book) is Finally Here! | 01-method, 02-discovery, 05-validation, 19-iteration | 持续发现, 机会解决方案树, 产品发现, 成果导向, 假设测试 | note | link |
| 2 | 9.55 | Finding participants for user research - Service Manual - GOV.UK | 02-discovery, 05-validation, 19-iteration, 01-method | 用户研究, 用户招募, 产品发现, 包容性设计, GOV.UK, 需求挖掘 | link | |
| 3 | 9.40 | How the discovery phase works - Service Manual - GOV.UK | 02-discovery, 01-method, 05-validation, 19-iteration | 敏捷方法论, 发现阶段, 需求挖掘, GOV.UK, 服务设计 | link | |
| 4 | 9.10 | Learning about users and their needs - Service Manual - GOV.UK | 02-discovery, 05-validation, 19-iteration, 01-method | 用户研究, 需求工程, GDS, 产品定义, 用户需求模板 | link | |
| 5 | 9.05 | Writing a user research plan - User Research Manual - Department for Education | 02-discovery, 05-validation, 19-iteration, 01-method | 用户研究, 研究计划, 需求挖掘, 产品验证, 设计规范 | link |
PRD & Specs（prd_spec）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.70 | OpenAPI Specification v3.2.0 | 03-prd, 07-engineering, 09-backend, 10-intelligence | API规范, OpenAPI, 契约驱动开发, 接口定义, Agent调用标准 | note | link |
| 2 | 9.55 | Overview | 03-prd, 07-engineering, 09-backend, 10-intelligence | API Linting, OpenAPI, API 治理, 契约校验, JSON/YAML | note | link |
UX / UI & Design Systems（ux_ui）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.85 | Web Content Accessibility Guidelines (WCAG) 2.2 | 04-prototype, 06-ui, 08-frontend, 05-validation | 无障碍, WCAG, W3C标准, UX设计, 前端开发, 合规性 | note | link |
| 2 | 9.55 | Good from Afar, But Far from Good: AI Prototyping in Real Design Contexts - NN/G | 04-prototype, 05-validation, 06-ui, 08-frontend | AI 原型, UX 设计, UI 生成, 设计工作流, 可用性测试 | link | |
| 3 | 9.40 | PAIR Guidebook | 04-prototype, 06-ui, 08-frontend, 05-validation | 人机交互, UX 设计, AI 心理学, 设计规范 | note | link |
| 4 | 9.40 | Designing AI Products and Features: Study Guide - NN/G | 04-prototype, 06-ui, 08-frontend, 05-validation | AI产品设计, UX/UI, 价值主张, 提示词辅助, 生成式UI, NN/g | link | |
| 5 | 9.25 | Guidelines for Human-AI Interaction - Microsoft Research | 04-prototype, 06-ui, 05-validation, 08-frontend | 人机交互, AI设计准则, 用户体验, HAX Toolkit, 负责任AI | note | link |
| 6 | 9.20 | Design Tokens Format Module 2025.10 | 04-prototype, 06-ui, 08-frontend, 05-validation | 设计令牌, 设计系统, W3C规范草案, JSON交换格式, UI工程化 | note | link |
| 7 | 9.20 | axe-core | 04-prototype, 06-ui, 08-frontend, 05-validation | 可访问性, 自动化测试, WCAG, 前端开发, UI | note | link |
Engineering Workflow（engineering）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.35 | [OWASP Application Security Verification Standard (ASVS) | OWASP Foundation](sources/md/owasp-org-www-project-application-security-verification-standard-50b4bc9ed328.md) | 07-engineering, 09-backend, 08-frontend, 17-deployment | Web安全, 安全标准, OWASP, 安全开发, 渗透测试 | note |
| 2 | 9.20 | [Fast and reliable end-to-end testing for modern web apps | Playwright](sources/md/playwright-dev-8960df5ea200.md) | 07-engineering, 08-frontend, 09-backend, 17-deployment | 端到端测试, 自动化测试, Playwright, 软件工程, 前端开发 | note |
| 3 | 9.20 | [Test runner | Storybook docs](sources/md/storybook-js-org-docs-writing-tests-integrations-test-runner-28169fa9753f.md) | 07-engineering, 08-frontend, 09-backend, 17-deployment | Storybook, 自动化测试, Playwright, 组件驱动开发, CI/CD | note |
| 4 | 9.05 | Set up a service team at each phase - Service Manual - GOV.UK | 07-engineering, 08-frontend, 09-backend, 17-deployment | 敏捷开发, 多学科团队, 团队角色, GDS, 产品生命周期, 交付管理 | link |
Backend & Reliability（backend）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.20 | RFC 9457: Problem Details for HTTP APIs | 09-backend, 07-engineering, 17-deployment, 18-evaluation | HTTP API, 错误处理, 标准规范, RFC 9457, JSON | note | link |
RAG（rag）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.10 | [2005.11401] Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks | 10-intelligence, 13-data, 18-evaluation, 07-engineering | RAG, 检索增强, DPR, BART, 知识图谱, 深度学习, 事实准确性 | note | link |
Agents & Tool Use（agents）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | [2308.08155] AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation | 07-engineering, 10-intelligence, 18-evaluation, 01-method | 多智能体, LLM框架, 对话编程, AutoGen, 人机协作 | note | link |
| 2 | 9.40 | LangGraph overview - Docs by LangChain | 07-engineering, 10-intelligence, 18-evaluation, 01-method | Agent 编排, 有状态工作流, LangGraph, LangChain, 人机协作 | note | link |
| 3 | 9.25 | [2210.03629] ReAct: Synergizing Reasoning and Acting in Language Models | 07-engineering, 10-intelligence, 18-evaluation, 01-method | ReAct, 智能体, 思维链, 交互式决策, 推理与行动 | note | link |
Data（data）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.25 | [1803.09010] Datasheets for Datasets | 13-data, 20-governance, 18-evaluation, 10-intelligence | 数据集文档, 数据治理, 透明度, 机器学习偏见, 数据管理 | note | link |
Training & Alignment（training）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.70 | [2305.18290] Direct Preference Optimization: Your Language Model is Secretly a Reward Model | 14-pretrain, 15-posttrain, 18-evaluation, 13-data | DPO, RLHF, 模型对齐, LLM, 微调 | note | link |
| 2 | 9.55 | [2106.09685] LoRA: Low-Rank Adaptation of Large Language Models | 14-pretrain, 15-posttrain, 18-evaluation, 13-data | LoRA, PEFT, 大模型微调, 参数高效, 模型压缩 | note | link |
| 3 | 9.55 | [2305.14314] QLoRA: Efficient Finetuning of Quantized LLMs | 14-pretrain, 15-posttrain, 18-evaluation, 13-data | QLoRA, 低比特微调, 显存优化, 量化技术, Guanaco | note | link |
| 4 | 9.40 | [2212.10560] Self-Instruct: Aligning Language Models with Self-Generated Instructions | 14-pretrain, 13-data, 15-posttrain, 18-evaluation | Self-Instruct, 合成数据, 指令微调, 模型对齐, 数据管道 | note | link |
| 5 | 9.10 | [2203.02155] Training language models to follow instructions with human feedback | 14-pretrain, 15-posttrain, 18-evaluation, 13-data | RLHF, InstructGPT, 模型对齐, PPO, 指令遵循 | note | link |
Inference & Optimization（inference）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | [2309.06180] Efficient Memory Management for Large Language Model Serving with PagedAttention | 16-inference, 17-deployment, 18-evaluation, 07-engineering | vLLM, PagedAttention, KV Cache, 推理优化, 内存管理 | note | link |
| 2 | 9.40 | github-com-nvidia-tensorrt-llm | 16-inference, 17-deployment, 18-evaluation, 07-engineering | LLM推理, GPU加速, 量化, NVIDIA, 深度学习框架 | link | |
| 3 | 9.40 | [2210.17323] GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers | 16-inference, 17-deployment, 18-evaluation, 07-engineering | 模型量化, GPTQ, LLM部署, 推理加速, 后训练量化 | note | link |
| 4 | 9.40 | KServe | 16-inference, 17-deployment, 18-evaluation, 07-engineering | AI推理, Kubernetes, 模型服务, LLM运维, CNCF | note | link |
| 5 | 9.25 | Text Generation Inference | 16-inference, 17-deployment, 18-evaluation, 07-engineering | LLM 推理, 模型部署, Hugging Face, 量化优化, 高并发 | note | link |
Deployment & Operations（ops）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | Google SRE - Defining slo: service level objective meaning | 17-deployment, 18-evaluation, 07-engineering, 20-governance | SRE, SLO, SLI, 系统可靠性, 错误预算, 可观测性 | note | link |
| 2 | 9.50 | [Grafana OSS and Enterprise | Grafana documentation](sources/md/grafana-com-docs-grafana-latest-02ef4859d21b.md) | 17-deployment, 18-evaluation, 07-engineering, 20-governance | 可观测性, 监控, 可视化, 运维, 指标分析 | note |
| 3 | 9.35 | [Documentation | OpenTelemetry](sources/md/opentelemetry-io-docs-8de8f872c85e.md) | 17-deployment, 07-engineering, 18-evaluation, 20-governance | 可观测性, 云原生, OpenTelemetry, 分布式追踪, GenAI监控 | note |
Evaluation（evaluation）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.40 | [2306.05685] Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena | 18-evaluation, 17-deployment, 10-intelligence, 07-engineering | LLM-as-a-Judge, MT-bench, Chatbot Arena, 模型评测, 人类偏好对齐 | note | link |
| 2 | 9.20 | [Lighthouse | Chrome for Developers](sources/md/developer-chrome-com-docs-lighthouse-1b45ffb23796.md) | 17-deployment, 18-evaluation, 10-intelligence, 07-engineering | 网页审计, 前端性能, 可访问性, SEO, 自动化测试 | note |
| 3 | 9.20 | [2309.15217] Ragas: Automated Evaluation of Retrieval Augmented Generation | 18-evaluation, 10-intelligence, 17-deployment, 07-engineering | RAG, 自动化评估, LLM, Ragas, 检索增强生成 | note | link |
User / Auth / Audit（user_auth）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | Final: OpenID Connect Core 1.0 incorporating errata set 2 | 09-backend, 11-user, 20-governance, 07-engineering | OIDC, OAuth 2.0, 身份认证, ID Token, 用户声明 | note | link |
| 2 | 9.25 | RFC 7519: JSON Web Token (JWT) | 09-backend, 11-user, 20-governance, 07-engineering | JWT, RFC, 身份认证, API安全, JSON | note | link |
| 3 | 9.10 | RFC 6749: The OAuth 2.0 Authorization Framework | 09-backend, 11-user, 20-governance, 07-engineering | OAuth 2.0, 授权框架, 身份认证, API 安全, RFC 标准 | note | link |
Billing & Pricing（billing）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.25 | [従量課金 | Stripe のドキュメント](sources/md/docs-stripe-com-billing-subscriptions-usage-based-06655737082c.md) | 12-billing, 19-iteration, 20-governance, 09-backend | Stripe, 按量计费, SaaS, 支付系统, Token计费 | note |
| 2 | 9.20 | PCI Security Standards Council – Protect Payment Data with Industry-driven Security Standards, Training, and Programs | 12-billing, 20-governance, 19-iteration, 09-backend | PCI DSS, 支付安全, 合规性, 数据保护, 金融标准 | note | link |
Governance & Security（governance）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | [AI Risk Management Framework | NIST](sources/md/www-nist-gov-itl-ai-risk-management-framework-54dbc1541342.md) | 20-governance, 11-user, 12-billing, 13-data | AI治理, 风险管理, NIST标准, 可信AI, 合规性 | note |
General References（general）¶
| # | Score | Title | Chapters | Tags | Notes | Source |
|---|---|---|---|---|---|---|
| 1 | 9.55 | [Overview | OpenAI Platform](sources/md/platform-openai-com-docs-00a8ca075d5c.md) | 01-method | OpenAI, GPT-5.2, 智能体, 模型优化, API文档 | note |
| 2 | 9.05 | Agile delivery - Service Manual - GOV.UK | 01-method | 敏捷开发, 项目管理, 政府服务设计, 交付流程, 用户故事 | link |