Back to Details

literature-review-sprint

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Literature Review Sprint

文献综述冲刺

Purpose

目的

Turn "I need to read more papers" into a bounded, systematic review. This skill follows the handbook's literature review advice: search systematically, prioritize high-quality sources, assess institution/year/venue/impact, summarize key ideas and limitations, and stay current with top conferences and recent papers.
The output is a usable map of the field, not a long undigested bibliography.
将“我需要读更多论文”转化为一场有边界、系统化的综述。本技能遵循手册中的文献综述建议：系统化搜索、优先选择高质量来源、评估机构/年份/会议/影响力、总结核心观点与局限性，以及跟进顶级会议和最新论文。
产出的成果是一份实用的领域图谱，而非冗长且未消化的参考文献列表。

When to Use

使用场景

  • User is starting a new topic
  • User wants to verify novelty of an idea
  • User is writing related work
  • User has many papers and no structure
  • User needs a reading plan before an advisor or group meeting
  • 用户开启新研究主题
  • 用户想要验证想法的创新性
  • 用户撰写相关工作部分
  • 用户拥有大量论文但缺乏结构化梳理
  • 用户需要在与导师或小组会议前制定阅读计划

Workflow

工作流程

Stage 1: Define the Review Question

阶段1：明确综述问题

Ask for:
  • Topic or problem
  • Target field or venue family
  • Purpose: novelty check, related work writing, method comparison, dataset search, or general catch-up
  • Deadline and available reading time
  • Existing seed papers, if any
Force a narrow review question. "Diffusion models" is too broad; "diffusion policies for robot manipulation under distribution shift" is workable.
询问以下信息：
  • 研究主题或问题
  • 目标领域或会议体系
  • 综述目的：创新性验证、相关工作撰写、方法对比、数据集搜索或领域前沿跟进
  • 截止日期与可用阅读时间
  • 现有种子论文（如有）
要将综述问题限定在狭窄范围内。“Diffusion models”过于宽泛；“diffusion policies for robot manipulation under distribution shift”才是可行的问题。

Stage 2: Build the Search Plan

阶段2：制定搜索计划

Create search queries for:
  • Google Scholar / Semantic Scholar
  • arXiv
  • Top venue proceedings in the field
  • Hugging Face Papers or similar recent-paper hubs for ML topics
  • References and citations of 2-3 seed papers
For current topics, prioritize the last 2 years first, then trace backward to canonical papers.
为以下平台创建搜索查询：
  • Google Scholar / Semantic Scholar
  • arXiv
  • 领域内顶级会议论文集
  • 针对机器学习主题的Hugging Face Papers或类似最新论文 hub
  • 2-3篇种子论文的参考文献与引用文献
对于前沿主题，优先搜索近2年的论文，再回溯经典文献。

Stage 3: Triage Papers

阶段3：论文筛选

Classify papers into:
  • Core
    : must read carefully
  • Context
    : skim for framing or related work
  • Maybe
    : keep if time remains
  • Reject
    : irrelevant, low quality, or superseded
Assess each paper using:
  • Institution or lab
  • Year
  • Venue
  • Citation/impact signal
  • Method relevance
  • Evaluation relevance
  • Clear limitation or gap
将论文分为以下类别：
  • 核心
    ：必须仔细阅读
  • 背景
    ：略读以了解框架或相关工作
  • 备选
    ：时间充裕时阅读
  • 排除
    ：无关、低质量或已被替代
通过以下维度评估每篇论文：
  • 所属机构或实验室
  • 发表年份
  • 会议/期刊
  • 引用量/影响力信号
  • 方法相关性
  • 评估相关性
  • 是否存在明确的局限性或研究空白

Stage 4: Extract Notes

阶段4：提取笔记

For each core paper, produce compact notes:
  • One-sentence contribution
  • Problem setting
  • Key method
  • Dataset/task
  • Baselines
  • Main result
  • Limitation
  • How it relates to the user's project
Keep notes short enough that they remain searchable.
针对每篇核心论文，生成简洁笔记：
  • 一句话总结核心贡献
  • 问题设定
  • 核心方法
  • 数据集/任务
  • 基线模型
  • 主要结果
  • 局限性
  • 与用户项目的关联
笔记需足够简短，以便于搜索查找。

Stage 5: Synthesize the Map

阶段5：合成领域图谱

Create a field map organized by axes that matter for the topic, such as:
  • Method families
  • Dataset/task families
  • Assumptions
  • Evaluation metrics
  • Failure modes
  • Open gaps
Name 3-5 concrete gaps or tensions. Separate real gaps from "I have not read enough yet."
根据主题相关维度构建领域图谱，例如：
  • 方法体系
  • 数据集/任务体系
  • 假设前提
  • 评估指标
  • 失效模式
  • 研究空白
列出3-5个具体的研究空白或矛盾点。区分真实空白与“我尚未阅读足够文献”的情况。

Stage 6: Produce the Artifact

阶段6：生成成果文件

Save to 
~/phd-log/literature/YYYY-MM-DD-[topic].md
.
markdown
undefined
保存至 
~/phd-log/literature/YYYY-MM-DD-[topic].md
。
markdown
undefined

Literature Review Sprint — [Topic]

Literature Review Sprint — [Topic]

Review question

Review question

[Specific question]
[Specific question]

Search plan

Search plan

  • Queries:
  • Venues / sources:
  • Seed papers:
  • Queries:
  • Venues / sources:
  • Seed papers:

Paper triage

Paper triage

TitleYearVenueCategoryWhy
TitleYearVenueCategoryWhy

Core paper notes

Core paper notes

[Paper title]

[Paper title]

  • Contribution:
  • Setting:
  • Method:
  • Evaluation:
  • Main result:
  • Limitation:
  • Relevance:
  • Contribution:
  • Setting:
  • Method:
  • Evaluation:
  • Main result:
  • Limitation:
  • Relevance:

Field map

Field map

  • Method families:
  • Evaluation setups:
  • Common assumptions:
  • Failure modes:
  • Method families:
  • Evaluation setups:
  • Common assumptions:
  • Failure modes:

Candidate gaps

Candidate gaps

  1. [gap + evidence]
  2. [gap + evidence]
  3. [gap + evidence]
  1. [gap + evidence]
  2. [gap + evidence]
  3. [gap + evidence]

Next actions

Next actions

  • Read:
  • Reproduce/check:
  • Ask:
  • Update idea/related work:
undefined
  • Read:
  • Reproduce/check:
  • Ask:
  • Update idea/related work:
undefined

Tone

语气

Be structured and skeptical. The user should leave with a smaller, clearer reading list and a sense of what matters.
保持结构化与批判性。用户结束后应得到一份更精简、清晰的阅读列表，并明确重点内容。

What Not to Do

禁忌事项

  • Do not produce generic paper summaries without synthesis.
  • Do not treat citation count as the only quality signal.
  • Do not over-prioritize old canonical papers when the user needs current trends.
  • Do not let the user claim novelty before checking recent top venues and arXiv.
  • 不要生成无整合的通用论文摘要。
  • 不要将引用量作为唯一的质量判断标准。
  • 当用户需要了解前沿趋势时，不要过度优先旧的经典文献。
  • 在未查阅最新顶级会议和arXiv论文前，不要让用户宣称想法具有创新性。