serper-scholar

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Google Scholar Search Tool

基于 Google Scholar API 的学术文献搜索工具，提供学术论文、研究报告、技术文献的专业搜索能力。

An academic literature search tool based on Google Scholar API, providing professional search capabilities for academic papers, research reports, and technical literature.

When to Activate

当用户提到以下内容时自动激活：

Automatically activate when the user mentions the following:

学术搜索关键词

Academic Search Keywords

"论文"、"学术"、"文献"、"研究"
"搜索论文"、"查找文献"、"学术研究"
"谷歌学术"、"Scholar"

"papers", "academic", "literature", "research"
"search papers", "find literature", "academic research"
"Google Scholar", "Scholar"

特定场景

Specific Scenarios

需要查找学术论文或研究报告
需要了解某领域的学术进展
需要查找特定作者的作品
需要获取引用信息和发表刊物
需要研究技术领域的理论依据

Need to find academic papers or research reports
Need to understand academic progress in a certain field
Need to find works by specific authors
Need to obtain citation information and publication journals
Need to research theoretical basis in technical fields

示例问题

Example Questions

"帮我搜索关于机器学习的论文"
"查找一下深度学习在 NLP 中的应用"
"研究一下 Transformer 架构的学术论文"
"找一些关于大模型训练方法的文献"
"搜索一下 Attention mechanism 的相关论文"

"Help me search for papers on machine learning"
"Find applications of deep learning in NLP"
"Research academic papers on Transformer architecture"
"Find some literature on large model training methods"
"Search for papers related to Attention mechanism"

Tools

serper_scholar

用途： 执行学术文献搜索，返回论文详细信息

参数：

```
query
```
(必选，string)：搜索关键词
```
num
```
(可选，number)：返回结果数量，默认 10，最大 20
```
gl
```
(可选，string)：国家代码，默认 cn
推荐值： cn（中国）、us（美国）、uk（英国）
```
hl
```
(可选，string)：语言代码，默认 zh-CN
推荐值： zh-CN（简体中文）、en（英文）

返回字段：

```
title
```
：论文标题
```
url
```
：论文链接
```
snippet
```
：摘要
```
type
```
：文献类型（PDF、HTML 等）
```
year
```
：发表年份
```
authors
```
：作者列表
```
publication
```
：发表刊物/会议
```
citationCount
```
：引用次数

Purpose: Perform academic literature search and return detailed paper information

Parameters:

```
query
```
(required, string): Search keywords
```
num
```
(optional, number): Number of returned results, default 10, maximum 20
```
gl
```
(optional, string): Country code, default cn
Recommended values: cn (China), us (United States), uk (United Kingdom)
```
hl
```
(optional, string): Language code, default zh-CN
Recommended values: zh-CN (Simplified Chinese), en (English)

Return Fields:

```
title
```
: Paper title
```
url
```
: Paper link
```
snippet
```
: Abstract
```
type
```
: Literature type (PDF, HTML, etc.)
```
year
```
: Publication year
```
authors
```
: Author list
```
publication
```
: Publication journal/conference
```
citationCount
```
: Citation count

Best Practices

1. 搜索技巧

1. Search Techniques

使用专业术语和技术关键词：

示例：

✅ "Attention mechanism neural machine translation"
✅ "Transformer large language models"
✅ "Reinforcement learning robotics"
❌ "机器学习"（太宽泛，结果太多）

Use professional terms and technical keywords:

Examples:

✅ "Attention mechanism neural machine translation"
✅ "Transformer large language models"
✅ "Reinforcement learning robotics"
❌ "Machine learning" (Too broad, too many results)

2. 添加领域限定

2. Add Field Limitations

明确研究领域和方法：

示例：

✅ "BERT semantic analysis NLP"
✅ "CNN image classification computer vision"
✅ "GPT text generation natural language"
✅ "Q-learning reinforcement learning agent"

Clearly define research fields and methods:

Examples:

✅ "BERT semantic analysis NLP"
✅ "CNN image classification computer vision"
✅ "GPT text generation natural language"
✅ "Q-learning reinforcement learning agent"

3. 时间范围搜索

3. Time Range Search

关注最新研究进展：

示例：

✅ "Large language models 2024 2025"
✅ "Transformer architecture recent advances"
✅ "Diffusion models 2023 2024"

Focus on the latest research progress:

Examples:

✅ "Large language models 2024 2025"
✅ "Transformer architecture recent advances"
✅ "Diffusion models 2023 2024"

4. 作者和机构搜索

4. Author and Institution Search

查找特定研究者或机构的工作：

示例：

✅ "Geoffrey Hinton deep learning"
✅ "Yann LeCun CNN papers"
✅ "Andrew Ng machine learning"
✅ "OpenAI research papers"

Find works by specific researchers or institutions:

Examples:

✅ "Geoffrey Hinton deep learning"
✅ "Yann LeCun CNN papers"
✅ "Andrew Ng machine learning"
✅ "OpenAI research papers"

5. 论文类型筛选

5. Paper Type Filtering

关注特定类型的文献：

示例：

✅ "Survey deep learning"
✅ "Review transformer models"
✅ "Tutorial reinforcement learning"
✅ "Benchmark NLP models"

Focus on specific types of literature:

Examples:

✅ "Survey deep learning"
✅ "Review transformer models"
✅ "Tutorial reinforcement learning"
✅ "Benchmark NLP models"

6. 结果数量选择

6. Result Quantity Selection

根据需求调整：

快速浏览：
```
num=5
```
（核心文献）
全面了解：
```
num=10
```
（主流研究）
深度调研：
```
num=20
```
（全面覆盖）

Adjust according to needs:

Quick browsing:
```
num=5
```
(core literature)
Comprehensive understanding:
```
num=10
```
(mainstream research)
In-depth research:
```
num=20
```
(full coverage)

7. 引用信息分析

7. Citation Information Analysis

关注高引用论文和经典文献：

关注点：

引用次数：
```
citationCount
```
高的论文通常是领域经典
发表年份：较新的论文代表最新进展
发表刊物：顶级会议（NeurIPS、ICML、ACL）质量高

Focus on highly cited papers and classic literature:

Key Points:

Citation count: Papers with high
```
citationCount
```
are usually field classics
Publication year: Newer papers represent the latest progress
Publication journal: Top conferences (NeurIPS, ICML, ACL) have high quality

Example Scenarios

场景 1：技术调研

Scenario 1: Technical Research

用户提问： "研究一下 Transformer 架构的论文"

AI 行为：

javascript

// 搜索核心论文
serper_scholar({
  query: "Transformer architecture attention is all you need",
  num: 5
})

// 搜索后续研究
serper_scholar({
  query: "BERT transformer NLP 2018 2019",
  num: 8
})

回复示例：

核心论文：

1. **Attention Is All You Need** (2017)
   - 作者：Vaswani et al.
   - 引用：100,000+
   - 📄 [PDF]
   
后续研究：

2. **BERT: Pre-training of Deep Bidirectional Transformers** (2019)
   - 作者：Devlin et al.
   - 引用：80,000+
   - 发表：NAACL 2019
   
3. **Improving Language Understanding with Generative Pre-Training** (2018)
   - 作者：Radford et al. (OpenAI)
   - 引用：50,000+

User Query: "Research papers on Transformer architecture"

AI Action:

javascript

// 搜索核心论文
serper_scholar({
  query: "Transformer architecture attention is all you need",
  num: 5
})

// 搜索后续研究
serper_scholar({
  query: "BERT transformer NLP 2018 2019",
  num: 8
})

Response Example:

Core Papers:

1. **Attention Is All You Need** (2017)
   - Authors: Vaswani et al.
   - Citations: 100,000+
   - 📄 [PDF]
   
Follow-up Research:

2. **BERT: Pre-training of Deep Bidirectional Transformers** (2019)
   - Authors: Devlin et al.
   - Citations: 80,000+
   - Published: NAACL 2019
   
3. **Improving Language Understanding with Generative Pre-Training** (2018)
   - Authors: Radford et al. (OpenAI)
   - Citations: 50,000+

场景 2：领域综述

Scenario 2: Field Review

用户提问： "找一些关于强化学习的综述论文"

AI 行为：

javascript

serper_scholar({
  query: "Survey review reinforcement learning",
  num: 8
})

回复示例：

强化学习综述论文：

1. **Reinforcement Learning: A Survey** (2021)
   - 引用：5,000+
   - 类型：综述
   
2. **Deep Reinforcement Learning: An Overview** (2020)
   - 引用：8,000+
   - 类型：综述
   
3. **Recent Advances in Deep Reinforcement Learning** (2022)
   - 引用：2,000+
   - 类型：综述

User Query: "Find some review papers on reinforcement learning"

AI Action:

javascript

serper_scholar({
  query: "Survey review reinforcement learning",
  num: 8
})

Response Example:

Reinforcement Learning Review Papers:

1. **Reinforcement Learning: A Survey** (2021)
   - Citations: 5,000+
   - Type: Review
   
2. **Deep Reinforcement Learning: An Overview** (2020)
   - Citations: 8,000+
   - Type: Review
   
3. **Recent Advances in Deep Reinforcement Learning** (2022)
   - Citations: 2,000+
   - Type: Review

场景 3：作者研究

Scenario 3: Author Research

用户提问： "查找 Geoffrey Hinton 的深度学习论文"

AI 行为：

javascript

serper_scholar({
  query: "Geoffrey Hinton deep learning backpropagation",
  num: 10
})

回复示例：

Geoffrey Hinton 的经典论文：

1. **Deep Learning** (2015)
   - 合作：LeCun, Bengio
   - 发表：Nature
   - 引用：50,000+
   
2. **Backpropagation Applied to Handwritten Zip Code Recognition** (1989)
   - 引用：10,000+
   
3. **Reducing the Dimensionality of Data with Neural Networks** (2006)
   - 引用：30,000+

User Query: "Find Geoffrey Hinton's deep learning papers"

AI Action:

javascript

serper_scholar({
  query: "Geoffrey Hinton deep learning backpropagation",
  num: 10
})

Response Example:

Geoffrey Hinton's Classic Papers:

1. **Deep Learning** (2015)
   - Co-authors: LeCun, Bengio
   - Published: Nature
   - Citations: 50,000+
   
2. **Backpropagation Applied to Handwritten Zip Code Recognition** (1989)
   - Citations: 10,000+
   
3. **Reducing the Dimensionality of Data with Neural Networks** (2006)
   - Citations: 30,000+

场景 4：最新进展

Scenario 4: Latest Progress

用户提问： "2024-2025 年大模型有什么新进展"

AI 行为：

javascript

serper_scholar({
  query: "Large language models 2024 2025 advances",
  num: 10
})

回复示例：

2024-2025 大模型研究进展：

1. **Scaling Laws for Neural Language Models** (2024)
   - 持续探索模型规模与性能关系
   
2. **Instruction Tuning for Large Language Models** (2024)
   - 指令微调方法改进
   
3. **Chain-of-Thought Prompting** (2024)
   - 推理链提示技术

User Query: "What are the new advances in large models in 2024-2025"

AI Action:

javascript

serper_scholar({
  query: "Large language models 2024 2025 advances",
  num: 10
})

Response Example:

2024-2025 Large Model Research Advances:

1. **Scaling Laws for Neural Language Models** (2024)
   - Continued exploration of the relationship between model scale and performance
   
2. **Instruction Tuning for Large Language Models** (2024)
   - Improvements in instruction tuning methods
   
3. **Chain-of-Thought Prompting** (2024)
   - Chain-of-thought prompting technology

场景 5：跨学科搜索

Scenario 5: Interdisciplinary Search

用户提问： "搜索机器学习在医疗诊断中的应用论文"

AI 行为：

javascript

serper_scholar({
  query: "Machine learning medical diagnosis healthcare",
  num: 8
})

回复示例：

医疗诊断中的机器学习：

1. **Deep Learning for Medical Image Analysis** (2021)
   - 引用：8,000+
   - 应用：影像诊断
   
2. **Machine Learning in Clinical Diagnosis** (2022)
   - 引用：3,000+
   - 应用：辅助诊断
   
3. **AI in Healthcare: A Survey** (2023)
   - 引用：2,000+
   - 类型：综述

User Query: "Search for papers on the application of machine learning in medical diagnosis"

AI Action:

javascript

serper_scholar({
  query: "Machine learning medical diagnosis healthcare",
  num: 8
})

Response Example:

Machine Learning in Medical Diagnosis:

1. **Deep Learning for Medical Image Analysis** (2021)
   - Citations: 8,000+
   - Application: Image diagnosis
   
2. **Machine Learning in Clinical Diagnosis** (2022)
   - Citations: 3,000+
   - Application: Auxiliary diagnosis
   
3. **AI in Healthcare: A Survey** (2023)
   - Citations: 2,000+
   - Type: Review

Limitations

搜索结果来源： Google Scholar，可能受地区影响
访问限制： 某些论文需要订阅或付费访问
结果数量： 最多 20 条
更新延迟： 最新论文可能需要一段时间才会被收录
语言偏好： 英文论文数量远多于中文

Search Result Source: Google Scholar, may be affected by region
Access Restrictions: Some papers require subscription or paid access
Result Quantity: Maximum 20 results
Update Delay: Latest papers may take time to be indexed
Language Preference: Number of English papers is much higher than Chinese

Configuration

环境变量配置

Environment Variable Configuration

编辑

~/.openclaw/gateway.env

：

bash

SERPER_API_KEY=your-api-key-here

Edit

~/.openclaw/gateway.env

bash

SERPER_API_KEY=your-api-key-here

获取 API Key

Get API Key

访问 https://serper.dev/ 注册并获取 API Key。

免费额度：每月 2,500 次调用（Web 和 Scholar 共享）。

Visit https://serper.dev/ to register and obtain API Key.

Free quota: 2,500 calls per month (shared between Web and Scholar).

Related Tools

serper_search： 普通网页搜索
web_fetch： 获取单个网页的详细内容

serper_search: General web search
web_fetch: Get detailed content of a single webpage

Tips

混合使用： 先用 serper_search 了解概念，再用 serper_scholar 深入研究
引用优先： 优先阅读高引用论文（通常是领域经典）
关注年份： 平衡经典文献和最新研究
追踪作者： 找到重要作者后，搜索其全部作品
PDF 访问： 尝试访问论文页面，寻找免费版本

Hybrid Use: First use serper_search to understand concepts, then use serper_scholar for in-depth research
Citation Priority: Prioritize reading highly cited papers (usually field classics)
Focus on Year: Balance classic literature and latest research
Track Authors: After finding important authors, search all their works
PDF Access: Try accessing the paper page to find free versions

Version History

v1.0 (2026-02-06)：初始版本，基础学术搜索功能
- 支持 Google Scholar API
- 提供论文详细信息（作者、年份、引用等）
- 集成 OpenClaw Skill 系统

💡 提示： 学术搜索时，尽量使用英文关键词，英文论文数量和质量通常更高。

v1.0 (2026-02-06): Initial version with basic academic search functionality
- Supports Google Scholar API
- Provides detailed paper information (authors, year, citations, etc.)
- Integrated with OpenClaw Skill system

💡 Tip: When conducting academic searches, try to use English keywords as the quantity and quality of English papers are usually higher.