systems-paper-writing
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseSystems Paper Writing: Paragraph-Level Structural Blueprint
系统类论文撰写：段落级结构蓝图
Fine-grained structural guidance for writing 10–12 page systems papers targeting top systems venues: OSDI, SOSP, ASPLOS, NSDI, and EuroSys. This skill provides page allocation per section, paragraph-level blueprints, and writing patterns distilled from authoritative guides and best-paper analysis.
针对OSDI、SOSP、ASPLOS、NSDI和EuroSys等顶级系统类会议，为撰写10-12页系统类论文提供精细化结构指导。本技能提供各部分页面分配方案、段落级蓝图，以及从权威指南和优秀论文分析中提炼的写作模式。
When to Use This Skill
何时使用本技能
| Scenario | Use This Skill | Use ml-paper-writing Instead |
|---|---|---|
| Structuring a 12-page OSDI/SOSP paper | ✅ | |
| Page budget and paragraph planning | ✅ | |
| Systems-specific evaluation structure | ✅ | |
| General ML paper writing philosophy | ✅ | |
| Citation verification workflow | ✅ | |
| LaTeX templates and formatting | ✅ | |
| NeurIPS/ICML/ICLR paper structure | ✅ |
Boundary: ml-paper-writing provides general writing philosophy, multi-venue templates, and citation verification. This skill focuses exclusively on paragraph-level structural blueprints for systems conferences.
| 场景 | 使用本技能 | 改用ml-paper-writing技能 |
|---|---|---|
| 构建12页OSDI/SOSP论文结构 | ✅ | |
| 页面预算与段落规划 | ✅ | |
| 系统类论文专属评估结构 | ✅ | |
| 通用机器学习论文写作理念 | ✅ | |
| 引用验证流程 | ✅ | |
| LaTeX模板与格式设置 | ✅ | |
| NeurIPS/ICML/ICLR论文结构 | ✅ |
边界说明：ml-paper-writing技能提供通用写作理念、多会议模板和引用验证功能。本技能仅专注于系统类会议论文的段落级结构蓝图。
Authoritative Sources
权威参考来源
This blueprint synthesizes guidance from established systems researchers:
- Levin & Redell — "How (and How Not) to Write a Good Systems Paper" (SOSP'83 PC Chairs, USENIX/ACM SIGOPS)
- Irene Zhang (MSR/UW) — "Hints on how to write an SOSP paper" (SOSP/OSDI PC)
- Gernot Heiser (UNSW, seL4) — Style Guide + Paper Writing Talk
- Timothy Roscoe (ETH Zürich) — "Writing reviews for systems conferences"
- Mike Dahlin (UT Austin/Google) — "Giving a Conference Talk"
- Yi Ding — "How to write good systems papers?"
- hzwer & DingXiaoH — WritingAIPaper (GitHub 1.3k+ stars)
Full citations and URLs: see references/section-blueprints.md.
本蓝图整合了资深系统领域研究者的指导内容：
- Levin & Redell — 《如何撰写（及避免错误撰写）优秀系统类论文》（SOSP'83程序委员会主席，USENIX/ACM SIGOPS）
- Irene Zhang（微软研究院/华盛顿大学） — 《撰写SOSP论文的提示》（SOSP/OSDI程序委员会成员）
- Gernot Heiser（新南威尔士大学，seL4项目） — 风格指南+论文写作讲座
- Timothy Roscoe（苏黎世联邦理工学院） — 《为系统类会议撰写审稿意见》
- Mike Dahlin（得克萨斯大学奥斯汀分校/谷歌） — 《做会议演讲》
- Yi Ding — 《如何撰写优秀的系统类论文？》
- hzwer & DingXiaoH — WritingAIPaper（GitHub获1.3k+星标）
完整引用及链接：详见references/section-blueprints.md。
12-Page Systems Paper Blueprint
12页系统类论文蓝图
Overview: Page Allocation
概述：页面分配
| Section | Pages | Purpose |
|---|---|---|
| Abstract | ~0.25 | 150–250 words, 5-sentence structure |
| S1 Introduction | 1.5–2 | Problem → Gap → Insight → Contributions |
| S2 Background & Motivation | 1–1.5 | Terms + Production observations |
| S3 Design | 3–4 | Architecture + Module details + Alternatives |
| S4 Implementation | 0.5–1 | Prototype details, LOC, key engineering |
| S5 Evaluation | 3–4 | Setup + End-to-end + Microbenchmarks + Scalability |
| S6 Related Work | 1 | Grouped by methodology, explicit comparison |
| S7 Conclusion | 0.5 | 3-sentence summary |
| Total | ~12 | Submission: 12 pages strict (USENIX) / 11 pages (ACM ASPLOS). Camera-ready: up to 14 pages (USENIX) / 13 pages (ACM). Ranges above span submission through camera-ready. Target 12 pages for initial submission. References unlimited. |
| 章节 | 页数 | 目标 |
|---|---|---|
| 摘要 | ~0.25 | 150-250词，5句结构 |
| S1 引言 | 1.5–2 | 问题→研究缺口→核心洞见→贡献 |
| S2 背景与动机 | 1–1.5 | 术语定义+生产环境观察 |
| S3 设计 | 3–4 | 架构+模块细节+备选方案 |
| S4 实现 | 0.5–1 | 原型细节、代码行数、关键工程实现 |
| S5 评估 | 3–4 | 实验设置+端到端测试+微基准测试+可扩展性 |
| S6 相关工作 | 1 | 按方法论分组，明确对比 |
| S7 结论 | 0.5 | 3句总结结构 |
| 总计 | ~12 | 投稿阶段：USENIX会议严格12页 / ACM ASPLOS为11页。终稿阶段：USENIX会议最多14页 / ACM会议最多13页。上述页数范围覆盖投稿至终稿阶段，初始投稿目标为12页。参考文献页数不限。 |
Abstract (150–250 words, 5 sentences)
摘要（150-250词，5句结构）
text
Sentence 1: Problem context and importance
Sentence 2: Gap in existing approaches
Sentence 3: Key insight or thesis ("X is better for Y in environment Z")
Sentence 4: Summary of approach and key results
Sentence 5: Broader impact or availabilitySource: Levin & Redell — "Can you state the new idea concisely? Use them in the abstract." Irene Zhang — "The abstract is harder to write because you cannot use terms or concepts you introduced in the paper."
text
Sentence 1: Problem context and importance
Sentence 2: Gap in existing approaches
Sentence 3: Key insight or thesis ("X is better for Y in environment Z")
Sentence 4: Summary of approach and key results
Sentence 5: Broader impact or availability参考来源：Levin & Redell — "你能否简洁表述新观点？将其用于摘要中。" Irene Zhang — "摘要撰写难度更高，因为你不能使用论文后续才介绍的术语或概念。"
S1 Introduction (1.5–2 pages)
S1 引言（1.5–2页）
Paragraph structure:
- Problem statement (~0.5 page) — Establish the domain and why it matters. Use concrete numbers (cluster sizes, workload statistics, latency requirements).
- Gap analysis (~0.5 page) — Enumerate specific gaps G1–Gn in existing systems. Each gap is one sentence with evidence.
- Key insight (1 paragraph) — The thesis statement: "X is better for applications Y running in environment Z." (Irene Zhang formula)
- Contributions (~0.5 page) — Numbered list of 3–5 concrete contributions. Each contribution is testable and maps to a section.
Writing pattern: hzwer Move 1 (Establish territory) → Move 2 (Find niche) → Move 3 (Occupy niche).
Source: Irene Zhang — "clearly state your target environment (Z) and application (Y)" + "clearly state why previous systems do not meet the needs"; Levin & Redell — "What exactly is the problem being solved?"
段落结构:
- 问题陈述（~0.5页）—— 确立研究领域及其重要性，使用具体数据（集群规模、工作负载统计、延迟要求）。
- 缺口分析（~0.5页）—— 列举现有系统中存在的具体缺口G1–Gn，每个缺口为带证据的单句。
- 核心洞见（1段）—— 核心论点：“在Z环境下，X更适合Y类应用”。（Irene Zhang提出的公式）
- 贡献说明（~0.5页）—— 3-5项具体贡献的编号列表，每项贡献可验证且对应论文某一章节。
写作模式：hzwer提出的三步法（确立研究领域→找到研究空白→填补研究空白）。
参考来源：Irene Zhang — "明确说明你的目标环境(Z)和应用(Y)" + "明确说明为何现有系统无法满足需求"; Levin & Redell — "要解决的具体问题是什么？"
S2 Background & Motivation (1–1.5 pages)
S2 背景与动机（1–1.5页）
Paragraph structure:
- Technical background (~0.5 page) — Define terms and systems the reader needs. Follow Gernot Heiser's "define-before-use" principle.
- Production observations (~0.5–1 page) — Present Observation 1, 2, 3 from real data or measurements. Each observation leads to a design insight.
Source: Irene Zhang — "clearly motivate Y and Z. Why is application Y important?"; Gernot Heiser — "define-before-use."
段落结构:
- 技术背景（~0.5页）—— 定义读者需了解的术语和系统，遵循Gernot Heiser提出的“先定义后使用”原则。
- 生产环境观察（~0.5–1页）—— 展示来自真实数据或测量的观察结果Observation 1、2、3，每项观察引出一个设计洞见。
参考来源：Irene Zhang — "明确说明Y和Z的动机。Y类应用为何重要？"; Gernot Heiser — "先定义后使用。"
S3 Design (3–4 pages)
S3 设计（3–4页）
Paragraph structure:
- System architecture overview (~0.5 page) — Architecture diagram first (Yi Ding: "draw a picture first"). One-paragraph walkthrough of major components and data flow.
- Module-by-module design (~2–2.5 pages) — Each subsection: what the module does, the design choice made, alternatives considered, and why this choice wins.
- Design alternatives and trade-offs (~0.5–1 page) — For each major decision, explicitly discuss what was not chosen and why.
Source: Irene Zhang — "Every design choice made in X should be discussed with alternatives and the reasons for the choice"; Levin & Redell — "What were the alternatives considered at various points, and why were the choices made?"
段落结构:
- 系统架构概述（~0.5页）—— 先展示架构图（Yi Ding："先画图"），用一段文字介绍主要组件和数据流。
- 模块级设计（~2–2.5页）—— 每个子章节说明：模块功能、做出的设计选择、考虑过的备选方案，以及选择该方案的原因。
- 设计备选方案与权衡（~0.5–1页）—— 针对每个重大决策，明确讨论未选择的方案及其原因。
参考来源：Irene Zhang — "X中的每一项设计选择都应结合备选方案及选择原因进行讨论"; Levin & Redell — "在各个阶段考虑过哪些备选方案，为何做出最终选择？"
S4 Implementation (0.5–1 page)
S4 实现（0.5–1页）
- Prototype description — Language, framework, LOC, integration with existing systems.
- Key engineering decisions — Non-obvious implementation choices worth documenting.
Source: Levin & Redell — "Does the paper describe something that has actually been implemented?"; Irene Zhang — "explain how you constructed a prototype to test your hypothesis."
- 原型描述—— 使用的语言、框架、代码行数、与现有系统的集成方式。
- 关键工程决策—— 记录非显而易见的实现选择。
参考来源：Levin & Redell — "论文是否描述了已实际实现的内容？"; Irene Zhang — "说明你如何构建原型来验证假设。"
S5 Evaluation (3–4 pages)
S5 评估（3–4页）
Paragraph structure:
- Experimental setup (~0.5 page) — Hardware, baselines, workloads, metrics. Enough detail to reproduce.
- End-to-end comparison (~1–1.5 pages) — X vs baselines for application Y on environment Z. Main performance results.
- Microbenchmarks / Ablation (~1–1.5 pages) — Isolate each design decision's contribution. Ablation experiments decompose the gains.
- Scalability (~0.5 page) — Show behavior as problem size, cluster size, or load increases.
Critical rule (Irene Zhang): State every experimental conclusion three times:
- Section opening: hypothesis ("We expect X to outperform Y because...")
- Section closing: conclusion ("Results show X outperforms Y by Z%")
- Figure caption: evidence ("Figure N shows X achieves Z% better throughput than Y")
Two experiment types:
- Type 1: X vs baselines for Y on Z (end-to-end comparison)
- Type 2: Ablation — remove each design component to measure its individual impact
段落结构:
- 实验设置（~0.5页）—— 硬件、基准系统、工作负载、指标，提供足够细节以确保可复现。
- 端到端对比（~1–1.5页）—— 在Z环境下，针对Y类应用对比X与基准系统的性能，展示主要结果。
- 微基准测试/消融实验（~1–1.5页）—— 分离每项设计决策的贡献，通过消融实验分解性能提升的来源。
- 可扩展性（~0.5页）—— 展示问题规模、集群规模或负载增加时的系统表现。
关键规则（Irene Zhang）：每个实验结论需表述三次:
- 章节开头：假设（"我们预计X优于Y，因为..."）
- 章节结尾：结论（"结果显示X优于Y Z%"）
- 图表标题：证据（"图N显示X的吞吐量比Y高Z%"）
两种实验类型:
- 类型1：在Z环境下针对Y类应用对比X与基准系统（端到端对比）
- 类型2：消融实验——移除每个设计组件以测量其单独影响
S6 Related Work (1 page)
S6 相关工作（1页）
- Group by methodology or approach, not by individual papers.
- For each group: what they do, what limitation remains, how your work differs.
- Use a comparison table when comparing 4+ systems on specific dimensions.
Source: Levin & Redell — "Are comparisons with previous work clear and explicit?"; Irene Zhang — use comparison tables.
- 按方法论或研究方法分组，而非按单篇论文分组。
- 针对每组：说明他们的研究内容、存在的局限性、你的工作与之的差异。
- 当对比4个以上系统的特定维度时，使用对比表格。
参考来源：Levin & Redell — "与已有工作的对比是否清晰明确？"; Irene Zhang — 使用对比表格。
S7 Conclusion (0.5 page)
S7 结论（0.5页）
Three sentences (Irene Zhang formula):
- The hypothesis / problem addressed
- The solution approach
- The key result
三句结构（Irene Zhang提出的公式）:
- 所解决的假设/问题
- 解决方案思路
- 核心结果
Writing Patterns
写作模式
Four reusable patterns for structuring systems papers. See references/writing-patterns.md for detailed examples.
四种可复用的系统类论文结构模式。详见references/writing-patterns.md中的详细示例。
Pattern 1: Gap Analysis (Lucid, ASPLOS'23)
模式1：缺口分析（Lucid, ASPLOS'23）
Enumerate gaps G1–Gn in Introduction → map to answers A1–An in Design. Creates a clear contract with the reader.
在引言中列举缺口G1–Gn → 在设计章节中对应给出解决方案A1–An，与读者建立清晰的约定。
Pattern 2: Observation-Driven (GFS, arXiv 2025)
模式2：观察驱动（GFS, arXiv 2025）
Present production observations (O1–O3) in Motivation → derive design insights → build system around insights. Effective when you have real workload data.
在动机章节中展示生产环境观察结果(O1–O3) → 推导设计洞见 → 围绕洞见构建系统。当你拥有真实工作负载数据时，该模式非常有效。
Pattern 3: Contribution List (Blox, EuroSys'24; Sia, SOSP'23)
模式3：贡献列表（Blox, EuroSys'24; Sia, SOSP'23）
Numbered contributions in Introduction, each mapping to a section. Readers (and reviewers) can track claims through the paper.
在引言中列出编号的贡献项，每项对应论文的一个章节。读者（及审稿人）可追踪论文中的各项主张。
Pattern 4: Thesis Formula (Irene Zhang)
模式4：核心论点公式（Irene Zhang）
Structure the entire paper around: "X is better for applications Y running in environment Z." Introduction states it, Design explains how, Evaluation proves it.
围绕“在Z环境下，X更适合Y类应用”构建整篇论文。引言提出论点，设计章节解释实现方式，评估章节验证论点。
Conference Differences
会议差异
Warning: Venue rules change yearly. Always verify against the current year's CFP before submission.
| Venue | Format | Submission Limit | Camera-Ready | References |
|---|---|---|---|---|
| OSDI | USENIX | 12 pages | 14 pages | Unlimited |
| NSDI | USENIX | 12 pages | 14 pages | Unlimited |
| SOSP | ACM SIGOPS | 12 pages (tech content) | — | Unlimited |
| ASPLOS | ACM SIGPLAN | 11 pages | 13 pages | Unlimited |
| EuroSys | ACM | 12 pages | — | Unlimited |
Based on 2025/2026 CFPs. Verify current limits before submission.
注意：会议规则每年都会变化。投稿前务必查阅当年的征稿启事（CFP）。
| 会议 | 格式 | 投稿页数限制 | 终稿页数 | 参考文献 |
|---|---|---|---|---|
| OSDI | USENIX | 12页 | 14页 | 不限 |
| NSDI | USENIX | 12页 | 14页 | 不限 |
| SOSP | ACM SIGOPS | 12页（技术内容） | — | 不限 |
| ASPLOS | ACM SIGPLAN | 11页 | 13页 | 不限 |
| EuroSys | ACM | 12页 | — | 不限 |
以上信息基于2025/2026年的征稿启事，投稿前请核实当前限制。
Writing Philosophy
写作理念
Manage Reader State (Gernot Heiser)
管理读者认知状态（Gernot Heiser）
Treat the reader's cognitive load like an OS managing process state. Never introduce a concept without context. Never reference something defined later without a forward pointer.
将读者的认知负担视为操作系统管理进程状态。绝不要在无上下文的情况下引入概念，绝不要引用后续才定义的内容而不添加前置指引。
Six-Dimensional Quality (Levin & Redell)
六维质量评估（Levin & Redell）
Self-check against: Original Ideas, Reality (is it built?), Lessons (what did you learn?), Choices (alternatives discussed?), Context (related work fair?), Presentation (clear writing?).
从以下维度自我检查：原创性、真实性（是否已实现？）、经验教训（你学到了什么？）、决策选择（是否讨论了备选方案？）、研究背景（相关工作对比是否公平？）、表达呈现（写作是否清晰？）。
Page-One Figure (hzwer)
首页图表（hzwer）
Include a figure on the first page that captures the core idea. Reviewers form first impressions from the title, abstract, and page-one figure.
在第一页包含一张体现核心观点的图表。审稿人会从标题、摘要和首页图表形成第一印象。
Academic Integrity Requirements
学术诚信要求
Citation Discipline
引用规范
- Never generate citations from memory. Use ml-paper-writing's citation verification workflow (Semantic Scholar / DBLP / CrossRef APIs).
- Mark unverified references as .
[CITATION NEEDED]
- 绝不要凭记忆生成引用。使用ml-paper-writing技能的引用验证流程（Semantic Scholar / DBLP / CrossRef API）。
- 未验证的引用标记为。
[CITATION NEEDED]
Prohibition of Fabrication
禁止编造
- Do NOT fabricate production observations, traces, deployment experiences, or experimental results.
- Do NOT generate fake venue rules, paper metadata, or best-paper claims.
- Do NOT copy paragraph-level text from reference papers. This blueprint provides structural guidance, not copy-paste templates.
- 不得编造生产环境观察结果、跟踪数据、部署经验或实验结果。
- 不得编造会议规则、论文元数据或优秀论文相关主张。
- 不得从参考论文中复制段落级文本。本蓝图提供结构指导，而非可直接复制粘贴的模板。
LLM Disclosure
LLM使用披露
- Some venues require disclosure of substantial LLM use in writing or ideation. Check each venue's AI policy in the current CFP.
- 部分会议要求披露在写作或构思过程中大量使用大语言模型（LLM）的情况。请查阅各会议当年征稿启事中的AI政策。
Attribution
署名归因
- When structures are inspired by specific papers (e.g., Lucid's gap-analysis pattern), cite the inspiration.
- Cross-repository references (e.g., ARIS paper-slides structure) are attributed, not copied.
- 当结构灵感来自特定论文（如Lucid的缺口分析模式）时，需引用灵感来源。
- 跨仓库引用（如ARIS论文-幻灯片结构）需注明归因，而非直接复制。
Temporal Validity
时效性
- Venue rules (page limits, format, AI policies) change annually. All venue information in this skill is based on 2025/2026 CFPs. Always verify against the current year's CFP.
- 会议规则（页数限制、格式、AI政策）每年都会变化。本技能中的所有会议信息基于2025/2026年的征稿启事。务必查阅当年的征稿启事进行核实。
Workflow: Structuring a New Systems Paper
工作流程：构建新系统类论文结构
text
Step 1: Read this SKILL.md for page allocation overview
Step 2: Read references/section-blueprints.md for per-section paragraph templates
Step 3: Choose a writing pattern from references/writing-patterns.md
Step 4: Draft section by section following the blueprint
Step 5: Run the checklist from references/checklist.md before submission
Step 6: Use ml-paper-writing for citation verification and LaTeX formattingtext
Step 1: Read this SKILL.md for page allocation overview
Step 2: Read references/section-blueprints.md for per-section paragraph templates
Step 3: Choose a writing pattern from references/writing-patterns.md
Step 4: Draft section by section following the blueprint
Step 5: Run the checklist from references/checklist.md before submission
Step 6: Use ml-paper-writing for citation verification and LaTeX formattingQuick Checklist
快速检查清单
- Thesis statement follows "X is better for Y in Z" formula
- Introduction has numbered contributions (3–5)
- Each contribution maps to a paper section
- Design discusses alternatives for every major choice
- Every eval conclusion stated 3 times (hypothesis, result, caption)
- Related work grouped by methodology, not individual papers
- Page budget within venue limits
- All citations verified programmatically (no hallucinated references)
- 核心论点遵循“X更适合Z环境下的Y类应用”公式
- 引言包含编号的贡献项（3-5项）
- 每项贡献对应论文的一个章节
- 设计章节针对每个重大决策讨论了备选方案
- 每个评估结论表述了三次（假设、结果、图表标题）
- 相关工作按方法论分组，而非按单篇论文分组
- 页数符合会议限制
- 所有引用均已通过程序验证（无虚构引用）
Common Issues and Solutions
常见问题及解决方案
| Issue | Solution |
|---|---|
| Paper feels like a "feature list" | Restructure around thesis formula: X better for Y in Z |
| Evaluation lacks depth | Add ablation experiments isolating each design decision |
| Reviewers say "incremental" | Strengthen gap analysis: make G1–Gn crisper with evidence |
| Design section too long | Move implementation details to S4, keep S3 at design level |
| Motivation feels weak | Add production observations with concrete numbers |
| Related work reads like a bibliography | Group by approach, add explicit differentiation |
| 问题 | 解决方案 |
|---|---|
| 论文看起来像“功能列表” | 围绕核心论点公式重构：X更适合Z环境下的Y类应用 |
| 评估内容缺乏深度 | 添加消融实验，分离每项设计决策的贡献 |
| 审稿人认为“增量贡献” | 强化缺口分析：用证据让G1–Gn更清晰明确 |
| 设计章节过长 | 将实现细节移至S4，S3仅保留设计层面内容 |
| 动机不足 | 添加带具体数据的生产环境观察结果 |
| 相关工作读起来像参考文献列表 | 按研究方法分组，添加明确的差异化说明 |
References
参考文献
Writing Guidance
写作指导
- references/section-blueprints.md — Detailed per-section paragraph templates with authoritative source quotes and best-paper structural examples
- references/writing-patterns.md — Four writing patterns with concrete paper examples
- references/section-blueprints.md — 详细的章节级段落模板，包含权威来源引用和优秀论文结构示例
- references/writing-patterns.md — 四种写作模式及具体论文示例
Venue-Specific
特定会议资料
- references/checklist.md — 7-stage pre-submission checklist covering structure, writing quality, evaluation rigor, design quality, academic integrity, venue-specific requirements (OSDI/NSDI/ASPLOS/SOSP/EuroSys), and final pass
- references/systems-conferences.md — Conference overview, deadlines, track descriptions, formatting requirements, submission rules, and format conversion guides
- references/reviewer-guidelines.md — How systems conference reviewers evaluate papers, with venue-specific criteria and common concerns
- references/checklist.md — 7阶段投稿前检查清单，涵盖结构、写作质量、评估严谨性、设计质量、学术诚信、特定会议要求（OSDI/NSDI/ASPLOS/SOSP/EuroSys）及最终审核
- references/systems-conferences.md — 会议概述、截止日期、赛道描述、格式要求、投稿规则及格式转换指南
- references/reviewer-guidelines.md — 系统类会议审稿人如何评估论文，包含特定会议标准及常见关注点
LaTeX Templates
LaTeX模板
- templates/osdi2026/ — OSDI 2026 (USENIX format)
- templates/nsdi2027/ — NSDI 2027 (USENIX format)
- templates/asplos2027/ — ASPLOS 2027 (ACM SIGPLAN format)
- templates/sosp2026/ — SOSP 2026 (ACM SIGPLAN format)
- templates/osdi2026/ — OSDI 2026（USENIX格式）
- templates/nsdi2027/ — NSDI 2027（USENIX格式）
- templates/asplos2027/ — ASPLOS 2027（ACM SIGPLAN格式）
- templates/sosp2026/ — SOSP 2026（ACM SIGPLAN格式）