Skip to content

windrise/windrise.github.io

BranchesTags

Repository files navigation

Xueming Fu - Personal Website

Personal academic website powered by Hugo and the Hugo Academic (Wowchemy) theme, featuring an automated paper collection and management system.

Overview

This is a professional academic website featuring:

  • Personal biography and research interests
  • Publications showcase
  • Project portfolio
  • Research experience timeline
  • Skills and expertise
  • Contact information
  • Automated Paper Collection System - AI-powered paper discovery, filtering, and management

Quick Start

Prerequisites

  • Hugo Extended (v0.110.0 or later)
  • Git

Local Development

  1. Clone the repository:
git clone https://github.com/windrise/windrise.github.io.git
cd windrise.github.io
  1. Start the Hugo development server:
hugo server -D
  1. Open your browser and visit http://localhost:1313

Customization Guide

1. Update Personal Information

Edit content/authors/admin/_index.md to update:

  • Your name and title
  • Bio and research interests
  • Education background
  • Social media links
  • Email and contact info

2. Add Your Publications

Create new publication entries in content/publication/:

hugo new content/publication/my-paper/index.md

Edit the generated file with your publication details:

  • Title, authors, date
  • Journal/conference name
  • Abstract and summary
  • PDF, code, and other links

3. Update Projects

Add project entries in content/project/:

hugo new content/project/my-project/index.md

4. Modify Experience

Edit content/home/experience.md to add your:

  • Education history
  • Work experience
  • Research positions
  • Internships

5. Update Skills

Edit content/home/skills.md to showcase your:

  • Technical skills
  • Programming languages
  • Tools and frameworks
  • Research expertise

6. Change Theme and Colors

Edit config/_default/params.yaml to customize:

  • Theme (day/night mode)
  • Font family and size
  • Color scheme
  • Layout options

7. Update Navigation Menu

Edit config/_default/menus.yaml to modify the navigation bar links.

8. Add Your Photo

Replace the avatar image:

static/img/avatar.jpg

Deployment

GitHub Pages

  1. Push your changes to GitHub:
git add .
git commit -m "Update personal website"
git push origin main

  1. Enable GitHub Pages in your repository settings:

    • Go to Settings → Pages
    • Set source to "GitHub Actions" or "main branch"

  2. Your site will be available at: https://windrise.github.io

Netlify

  1. Connect your GitHub repository to Netlify
  2. Set build command: hugo --gc --minify
  3. Set publish directory: public
  4. Deploy!

File Structure

.
├── config/          # Site configuration
│   └── _default/
│       ├── config.yaml   # Main config
│       ├── params.yaml   # Theme parameters
│       ├── menus.yaml    # Navigation menus
│       └── languages.yaml
├── content/         # Your content
│   ├── authors/     # Author profiles
│   ├── home/        # Homepage widgets
│   ├── publication/ # Publications
│   └── project/     # Projects
├── static/          # Static files
│   └── img/         # Images
└── themes/          # Hugo themes

Tips

  • Images: Add images to static/img/ or within specific content folders
  • SEO: Update meta descriptions in config/_default/params.yaml
  • Analytics: Add Google Analytics ID in params.yaml
  • Comments: Enable Disqus or other comment systems in params.yaml

Resources

License

This website is powered by the Hugo Academic theme.

Contact

For questions or suggestions, please contact: your.email@example.com

📊 Automated Paper Collection System - Implementation Status

🎯 Project Overview

An AI-powered system that automatically discovers, filters, and manages academic papers from arXiv, completely free and running on GitHub Actions.

✅ Completed Features (Weeks 1-2)

Week 1: Core Automation Pipeline

  • ✅ arXiv Scraper (scripts/arxiv_scraper.py)

    • Daily automated fetching from arXiv
    • Category filtering (cs.CV, cs.LG, eess.IV)
    • Keyword matching for research areas
    • Smart date-based querying

  • ✅ Smart Filter System (scripts/smart_filter.py)

    • Multi-dimensional scoring (field match, venue quality, citations, code, practicality)
    • Weighted ranking algorithm
    • Top N paper selection (default: 10 papers/day)
    • Detailed score breakdown

  • ✅ GitHub Actions Workflows

    • daily-paper-update.yml - Daily automated pipeline
    • process-approved-papers.yml - Process approved papers
    • hugo.yml - Website deployment

Week 2: AI Enhancement

  • ✅ Multi-API AI Summaries (scripts/generate_summaries_multi.py)

    • Support for 7+ API providers:
      • Google Gemini (Free) ⭐
      • Groq (Free) ⭐
      • DeepSeek (Cheap)
      • ZhipuAI (Free) ⭐
      • OpenAI GPT
      • Anthropic Claude
      • Moonshot Kimi
    • Auto-fallback mechanism
    • 3-5 sentence summaries
    • Key contributions extraction
    • Bilingual support ready

  • ✅ Audio Generation (scripts/generate_audio.py)

    • Text-to-speech using Edge TTS (Free)
    • Multiple voice options
    • MP3 format output

  • ✅ Review System (scripts/create_review_issue.py)

    • Auto-creates GitHub Issues for paper review
    • Beautiful markdown formatting
    • Label-based approval workflow
    • Paper metadata display

  • ✅ Paper Management (scripts/process_approved_papers.py)

    • Auto-categorization
    • YAML database integration
    • Duplicate prevention
    • Metadata management

  • ✅ Mindmap Visualization (scripts/generate_mindmap.py)

    • Auto-generate Mermaid.js mindmaps
    • Visual paper structure representation
    • Hugo shortcode integration
    • Interactive web display
    • Exports to markdown files

  • ✅ Citation Tracking System (scripts/citation_tracker.py)

    • Semantic Scholar API integration
    • Automatic citation count updates
    • Historical citation data tracking
    • Citation trend analysis
    • Impact metrics (influential citations)
    • Weekly automated updates
    • Citation reports generation

🚧 Remaining Tasks (Week 3+)

Priority 0: Papers Website Enhancement ✅ 核心功能已完成

参考网站：https://mrnerf.github.io/awesome-3D-gaussian-splatting/

文件： layouts/shortcodes/all-papers-enhanced.html, static/css/papers-dark-mode.css

阶段 1：基础 UI 改进 ✅ 完成

  • ✅ 改进论文卡片布局

    • 优化卡片样式和悬停效果（translateY(-4px)动画）
    • 响应式设计优化
    • 改进的徽章系统（分类、会议、类型、相关性）
    • 📌 待完成：添加论文缩略图支持（需要 arXiv API 或自动生成）

  • ✅ 摘要展开/折叠功能

    • "Show Abstract" / "Hide Abstract" 按钮
    • 平滑的展开/收起动画
    • 切换状态管理（data-expanded）

阶段 2：搜索和过滤系统 ✅ 完成

  • ✅ 实时搜索功能

    • 搜索框 UI（带清除按钮 ×）
    • 实时过滤标题、作者、摘要（基于 data-search-text）
    • 即时结果更新和计数

  • ✅ 高级过滤器

    • 年份过滤（动态生成所有年份）
    • 分类/标签多选过滤（带图标和颜色）
    • 快速过滤器（⭐ Starred、💻 Has Code、🏆 Foundation）
    • 所有过滤器组合工作

  • ✅ 排序功能

    • 按日期排序（新→旧 / 旧→新）
    • 按相关性评分排序
    • 按引用数排序
    • 按标题字母顺序排序（A-Z）

  • ✅ UI 反馈

    • 结果计数实时更新（"Showing X papers"）
    • 无结果提示页面
    • 重置所有过滤器按钮（🔄 Reset All）
    • 返回顶部按钮（↑，滚动 >300px 时显示）

阶段 3：高级交互功能 ✅ 选择功能完成

  • ✅ 深色模式支持

    • 自动检测系统主题（@media (prefers-color-scheme: dark)）
    • 所有组件完整适配深色模式（Slate 配色）
    • 优化的深色配色方案（背景 #1e293b，卡片 #0f172a）
    • ✅ 手动切换开关和 localStorage 主题持久化

  • ✅ UI 增强

    • 返回顶部按钮（带渐变背景和悬停动画）
    • 浮动操作按钮组（返回顶部 + 全选可见论文）
    • 📌 待完成：滚动进度指示器（顶部进度条）

  • ✅ 选择和分享模式（NEW! 刚刚实现）

    • ✅ 多选论文功能（checkbox，Selection Mode 切换）
    • ✅ 选中论文预览栏（固定在顶部，sticky positioning）
    • ✅ 生成分享链接（URL hash: #selected=id1,id2,id3）
    • ✅ 导出选中论文：
      • BibTeX 格式（.bib 文件）
      • JSON 格式（.json 文件）
      • Markdown 格式（.md 文件）
    • ✅ 分享模态框（复制链接到剪贴板）
    • ✅ 从 URL 加载选中状态（分享链接支持）
    • ✅ 全选可见论文功能

阶段 4：阅读笔记系统 ✅ Phase 1 & 2 已完成 (LocalStorage + Markdown + Autosave)

设计文档： docs/READING_NOTES_DESIGN.md

  • ✅ 笔记数据结构设计（NEW! 设计已完成）

    • 完整的 YAML schema 定义
    • 包含字段：status, priority, progress, rating, notes, highlights, tags, todos
    • 支持 Markdown 格式笔记
    • 笔记元数据（创建时间、更新时间、版本）
    • 笔记与论文关联

  • 📋 笔记编辑界面（计划中 - Phase 2）

    • 全屏/侧边栏笔记模态框
    • 集成 Markdown 编辑器（SimpleMDE 或类似）
    • 实时预览（Markdown → HTML）
    • 自动保存草稿（LocalStorage）
    • 格式化工具栏
    • 状态管理（to-read, reading, completed）
    • 进度追踪（0-100%）
    • 星级评分（1-5 星）

  • 📋 笔记展示（计划中 - Phase 3）

    • 论文卡片显示笔记指示器（badge）
    • 专门的笔记视图页面（/papers/notes/）
    • 笔记搜索和过滤（按标签、状态、评分）
    • 笔记时间线和统计
    • 阅读进度仪表板

  • 📋 笔记高级功能（计划中 - Phase 3-4）

    • 高亮系统（带颜色和注释）
    • 待办事项（To-Do）集成
    • 标签管理和标签云
    • 笔记导出（Markdown, PDF）
    • 笔记分享链接
    • 版本控制和历史记录
    • 笔记统计仪表板

实施方案：

  • Phase 1: 基础笔记（文本框 + 状态 + 评分）
  • Phase 2: Markdown 编辑器 + 自动保存
  • Phase 3: 高亮 + 待办 + 标签
  • Phase 4: 协作和高级功能

存储方案：

  • Hybrid 模式：LocalStorage（草稿）+ GitHub（持久化）
  • 自动保存：每 5 秒保存到 LocalStorage
  • 手动保存：提交到 papers.yaml via GitHub

阶段 5：可视化增强

  • 论文关系图

    • 基于引用的关系网络
    • 基于主题的聚类可视化
    • 交互式探索

  • 统计图表

    • 每月添加论文趋势
    • 研究领域分布饼图
    • 会议/期刊统计
    • 阅读进度追踪

  • 时间线视图

    • 论文时间线展示
    • 研究进展可视化
    • 里程碑标记

Priority 1: Essential Features

  • Mindmap Generation ✅ COMPLETED

    • ✅ Auto-generate paper structure visualization
    • ✅ Using Mermaid.js for web integration
    • ✅ Interactive expand/collapse
    • ✅ Export to Hugo pages
    • ✅ Hugo shortcode for easy embedding
    • Script: scripts/generate_mindmap.py

  • Citation Tracking ✅ COMPLETED

    • ✅ Integration with Semantic Scholar API (Free)
    • ✅ Auto-update citation counts
    • ✅ Citation history tracking
    • ✅ Weekly update schedule (GitHub Actions)
    • ✅ Impact tracking over time
    • ✅ Citation reports generation
    • Script: scripts/citation_tracker.py

Priority 2: Advanced Features

  • Local Q&A System ✅ COMPLETED

    • ✅ ChromaDB for vector storage
    • ✅ Sentence Transformers for embeddings (100% local)
    • ✅ Paper content indexing with smart chunking
    • ✅ Web interface for queries (Flask-based)
    • ✅ Command-line query tool with interactive mode
    • ✅ Semantic search across all papers
    • ✅ Find similar papers functionality
    • ✅ Works on GitHub Codespaces (60 hours/month free)
    • Scripts: scripts/setup_vectordb.py, scripts/query_papers.py, scripts/web_qa.py
    • Documentation: docs/QA_SYSTEM_GUIDE.md

  • Enhanced Management Interface (Week 3, Day 20-21)

    • Web-based admin panel (Hugo Admin)
    • CLI tool improvements
    • Batch operations
    • Statistics dashboard

Priority 3: Nice-to-Have Features

  • Weekly/Monthly Summary Reports

    • Auto-generate research trend reports
    • Top papers of the week/month
    • Category breakdowns
    • Email/Slack notifications

  • Paper Recommendations

    • Based on your collection
    • Similar paper suggestions
    • Author tracking
    • Conference/journal tracking

  • Enhanced Visualizations

    • Research field timeline
    • Citation network graphs
    • Keyword trend analysis
    • Author collaboration networks

  • Mobile App

    • Progressive Web App (PWA)
    • Offline reading support
    • Push notifications
    • Audio playback

📝 Next Development Session TODO

Priority 1 & 2 features are complete! Next development priorities:

A. Frontend Features - Papers Enhancement ⭐ 前端优先

  1. 阅读笔记系统 - Phase 1 ✅ 已完成

    • ✅ 创建笔记模态框 UI
    • ✅ 添加基础文本编辑区
    • ✅ 实现状态选择器（to-read, reading, completed）
    • ✅ 添加星级评分组件
    • ✅ LocalStorage 保存/加载
    • ✅ 在论文卡片添加"笔记"按钮
    • ✅ 显示笔记状态 badge

  2. 阅读笔记系统 - Phase 2 🚧 部分完成

    • ✅ 集成 SimpleMDE Markdown 编辑器
    • 实时预览功能 (已通过编辑器支持)
    • ✅ 自动保存（每 5 秒）
    • 同步到 GitHub（目前支持导出 Markdown）

  3. 论文缩略图支持 (优先级：中)

    • 从 arXiv 提取第一页作为缩略图
    • 或使用基于分类的默认图标
    • 添加懒加载（Intersection Observer）

  4. 主题切换增强 (优先级：低)

    • 手动主题切换按钮
    • localStorage 主题持久化
    • 平滑的主题切换动画

  5. 可视化增强 (优先级：低)

    • 论文关系图（基于引用）
    • 统计图表（Chart.js）
    • 时间线视图

前端已完成功能：

  • ✅ 搜索和过滤系统（完整）
  • ✅ 排序功能（5种排序方式）
  • ✅ 深色模式（自动检测）
  • ✅ 选择和分享功能（URL分享 + 3种导出格式）
  • ✅ 摘要展开/折叠
  • ✅ 响应式设计
  • ✅ 阅读笔记系统设计文档

B. Backend Automation Features (Priority 3)

  1. Weekly/Monthly Summary Reports ✅ COMPLETED

    • ✅ Auto-generate research trend reports
    • ✅ Top papers of the week/month
    • ✅ Category distribution and statistics
    • ✅ Citation growth tracking
    • ✅ Research highlights and recommendations
    • ✅ Automated GitHub Actions workflows (weekly & monthly)
    • ✅ Markdown format reports
    • Script: scripts/generate_summary_report.py
    • Documentation: docs/SUMMARY_REPORTS_GUIDE.md
    • Future: Email/Slack notifications integration

  2. Paper Recommendations ✅ COMPLETED

    • ✅ Recommendation engine based on content similarity
    • ✅ Trending papers detection
    • ✅ Similar paper suggestions (using vector database)
    • ✅ Author tracking and statistics
    • ✅ Conference/journal tracking
    • ✅ Interest-based recommendations
    • ✅ Comprehensive recommendation reports
    • Script: scripts/paper_recommender.py

  3. Enhanced CLI Tools ✅ COMPLETED

    • ✅ Batch operations (star, tag, annotate)
    • ✅ Export to multiple formats (JSON, CSV, BibTeX, Markdown)
    • ✅ Database validation and cleanup
    • ✅ Collection statistics
    • ✅ Paper management utilities
    • Script: scripts/paper_manager.py

  4. Data Analysis & Visualization ✅ COMPLETED

    • ✅ Category distribution analysis
    • ✅ Timeline analysis (papers over time)
    • ✅ Venue/conference analysis
    • ✅ Citation statistics and trends
    • ✅ Author analysis
    • ✅ ASCII charts + Mermaid.js diagrams
    • ✅ Comprehensive analysis reports
    • Script: scripts/analyze_collection.py
    • Documentation: docs/ADVANCED_TOOLS_GUIDE.md

  5. LLM-Enhanced Q&A ✅ COMPLETED

    • ✅ Natural language question answering
    • ✅ Uses Gemini or ZhipuAI APIs (你已配置的 API)
    • ✅ Paper comparison with AI analysis
    • ✅ Auto-generate literature reviews
    • ✅ Interactive chat interface
    • ✅ Context-aware answers from your collection
    • ✅ Source citations with relevance scores
    • Script: scripts/llm_qa.py
    • Documentation: docs/LLM_QA_GUIDE.md

🔧 Quick Commands

# Run the full pipeline manually
./scripts/test_pipeline.sh

# Test API keys
./scripts/test_api.sh

# Scrape papers (test mode)
python scripts/arxiv_scraper.py --days 1 --max-results 20

# Filter papers
python scripts/smart_filter.py --top-n 10

# Generate summaries (auto-select API)
python scripts/generate_summaries_multi.py --provider auto

# Generate mindmaps for all papers
python scripts/generate_mindmap.py

# Generate mindmap for specific paper
python scripts/generate_mindmap.py --paper-id <paper-id>

# Update citation counts
python scripts/citation_tracker.py

# Generate citation report
python scripts/citation_tracker.py --report --output reports/citation_report.md

# Force update citations (ignore recent check)
python scripts/citation_tracker.py --force

# Setup Q&A vector database
python scripts/setup_vectordb.py

# Query papers (command line)
python scripts/query_papers.py -q "your question here"

# Find similar papers
python scripts/query_papers.py --similar <paper-id>

# Interactive query mode
python scripts/query_papers.py -i

# Start web interface for Q&A
python scripts/web_qa.py

# LLM-enhanced Q&A (requires API key)
python scripts/llm_qa.py -i  # Interactive chat mode
python scripts/llm_qa.py -q "your question"  # Ask specific question
python scripts/llm_qa.py --compare paper-1 paper-2  # Compare papers
python scripts/llm_qa.py --review "topic"  # Generate literature review

# Generate summary reports
python scripts/generate_summary_report.py --period week
python scripts/generate_summary_report.py --period month
python scripts/generate_summary_report.py --period both

# Paper recommendations
python scripts/paper_recommender.py --trending
python scripts/paper_recommender.py --recommend-similar <paper-id>
python scripts/paper_recommender.py --track-author "Author Name"
python scripts/paper_recommender.py --report

# Paper management (batch operations)
python scripts/paper_manager.py stats
python scripts/paper_manager.py validate
python scripts/paper_manager.py export output.json --format json
python scripts/paper_manager.py star paper-1 paper-2

# Collection analysis
python scripts/analyze_collection.py --output reports/analysis.md
python scripts/analyze_collection.py --analysis categories

# Create review issue
python scripts/create_review_issue.py

# Process approved papers (after labeling issue)
python scripts/process_approved_papers.py --issue-number 123

📚 Documentation

Detailed guides available in /docs/:

  • QUICK_START.md - Get started in 5 minutes
  • SETUP_GUIDE.md - Complete setup instructions
  • API_SETUP.md - API key configuration
  • PAPER_AUTOMATION_PLAN.md - Full automation strategy
  • PRIORITY1_FEATURES.md - Mindmap & Citation Tracking guide
  • QA_SYSTEM_GUIDE.md - Local Q&A System complete guide
  • LLM_QA_GUIDE.md - LLM-Enhanced Q&A with Gemini/ZhipuAI
  • SUMMARY_REPORTS_GUIDE.md - Weekly/Monthly Reports guide
  • ADVANCED_TOOLS_GUIDE.md - Recommendations, Management & Analysis tools
  • READING_NOTES_DESIGN.md - Reading notes system design
  • TROUBLESHOOTING.md - Common issues and solutions

🎯 Success Metrics

Current achievements:

  • ✅ 100% automated paper discovery
  • ✅ Zero-cost operation (all free APIs)
  • ✅ ~10 minutes daily review time
  • ✅ Multi-API fallback for reliability
  • ✅ Full GitHub integration

Target metrics:

  • 5-10 papers reviewed daily
  • 2-3 papers added to collection weekly
  • 100% uptime with GitHub Actions
  • <15 minutes daily maintenance

🆓 Cost Breakdown

Service Monthly Cost Usage
GitHub Actions $0 2000 min/month free
API Keys (Gemini/Groq/Zhipu) $0 Free tiers
Edge TTS Audio $0 Unlimited
GitHub Pages Hosting $0 Unlimited
Storage (Git) $0 Unlimited for text
TOTAL $0/month 🎉

🚀 Future Enhancements

Ideas for later:

  • Integration with Zotero/Mendeley
  • Automated literature review generation
  • Paper relationship graphs
  • Collaborative filtering with other researchers
  • RSS feed generation
  • Social media auto-posting
  • Conference deadline tracking

📌 Important Notes

  1. API Keys Required: Set up at least one free API key (Gemini, Groq, or ZhipuAI)
  2. GitHub Secrets: Add API keys to repository secrets
  3. Daily Review: Check GitHub Issues daily for new papers
  4. Label System: Use approved, rejected, starred labels
  5. Backup: All data in data/papers/papers.yaml is version controlled

For detailed implementation plans, see docs/PAPER_AUTOMATION_PLAN.md

About

No description, website, or topics provided.

Resources

Readme

License

View license
Activity

Stars

0 stars

Watchers

1 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

 
 
 

Contributors

Languages