skills/claude-scientific-skills
1
0
Fork 0
mirror of https://github.com/K-Dense-AI/claude-scientific-skills.git synced 2026-01-26 16:58:56 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #24 from backtrue/feature/add-nstc-support

Browse Source 
feat(research-grants): Add comprehensive Taiwan NSTC proposal support
This commit is contained in:
Timothy Kassis
2026-01-05 12:38:29 -08:00
committed by GitHub
parent 70375367d6 4427270a71
commit d243a12564
2 changed files with 756 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
scientific-skills/research-grants/SKILL.md
View File

@@ -1,6 +1,6 @@
--- ---
name: research-grants name: research-grants
description: "Write competitive research proposals for NSF, NIH, DOE, and DARPA. Agency-specific formatting, review criteria, budget preparation, broader impacts, significance statements, innovation narratives, and compliance with submission requirements." description: "Write competitive research proposals for NSF, NIH, DOE, DARPA, and Taiwan NSTC. Agency-specific formatting, review criteria, budget preparation, broader impacts, significance statements, innovation narratives, and compliance with submission requirements."
allowed-tools: [Read, Write, Edit, Bash] allowed-tools: [Read, Write, Edit, Bash]
license: MIT license license: MIT license
metadata: metadata:
@@ -11,14 +11,14 @@ metadata:
## Overview ## Overview
Research grant writing is the process of developing competitive funding proposals for federal agencies and foundations. Master agency-specific requirements, review criteria, narrative structure, budget preparation, and compliance for NSF (National Science Foundation), NIH (National Institutes of Health), DOE (Department of Energy), and DARPA (Defense Advanced Research Projects Agency) submissions. Research grant writing is the process of developing competitive funding proposals for federal agencies and foundations. Master agency-specific requirements, review criteria, narrative structure, budget preparation, and compliance for NSF (National Science Foundation), NIH (National Institutes of Health), DOE (Department of Energy), DARPA (Defense Advanced Research Projects Agency), and Taiwan's NSTC (National Science and Technology Council) submissions.
**Critical Principle: Grants are persuasive documents that must simultaneously demonstrate scientific rigor, innovation, feasibility, and broader impact.** Each agency has distinct priorities, review criteria, formatting requirements, and strategic goals that must be addressed. **Critical Principle: Grants are persuasive documents that must simultaneously demonstrate scientific rigor, innovation, feasibility, and broader impact.** Each agency has distinct priorities, review criteria, formatting requirements, and strategic goals that must be addressed.
## When to Use This Skill ## When to Use This Skill
This skill should be used when: This skill should be used when:
- Writing research proposals for NSF, NIH, DOE, or DARPA programs - Writing research proposals for NSF, NIH, DOE, DARPA, or NSTC programs
- Preparing project descriptions, specific aims, or technical narratives - Preparing project descriptions, specific aims, or technical narratives
- Developing broader impacts or significance statements - Developing broader impacts or significance statements
- Creating research timelines and milestone plans - Creating research timelines and milestone plans
@@ -114,6 +114,16 @@ For detailed guidance on creating schematics, refer to the scientific-schematics
- Teaming and collaboration often required - Teaming and collaboration often required
- Varies dramatically by program manager and BAA (Broad Agency Announcement) - Varies dramatically by program manager and BAA (Broad Agency Announcement)
### NSTC (National Science and Technology Council - Taiwan)
**Mission**: Advance scientific breakthrough, industrial application, and societal impact in Taiwan.
**Key Features**:
- **CM03 Form**: The core technical proposal format.
- **Bilingual**: Abstract required in both Chinese and English.
- **Innovation & Feasibility**: Primary review focus.
- **Preliminary Data**: Highly critical for credibility.
- **Research Architecture Diagram**: A mandatory visual element for clarity.
## Core Components of Research Proposals ## Core Components of Research Proposals
### 1. Executive Summary / Project Summary / Abstract ### 1. Executive Summary / Project Summary / Abstract
@@ -195,6 +205,7 @@ For detailed agency-specific guidance, refer to:
- `references/nih_guidelines.md` - `references/nih_guidelines.md`
- `references/doe_guidelines.md` - `references/doe_guidelines.md`
- `references/darpa_guidelines.md` - `references/darpa_guidelines.md`
- `references/nstc_guidelines.md`
### 3. Specific Aims (NIH) or Objectives (NSF/DOE/DARPA) ### 3. Specific Aims (NIH) or Objectives (NSF/DOE/DARPA)
@@ -582,12 +593,17 @@ Varies by program office, but generally includes:
**DARPA-specific considerations**: **DARPA-specific considerations**:
- Overall scientific and technical merit - Overall scientific and technical merit
- Potential contribution to DARPA mission - Potential contribution to DARPA mission
- Relevance to stated program goals
- Plans and capability to accomplish technology transition
- Qualifications and experience of proposed team
- Realism of proposed costs and availability of funds - Realism of proposed costs and availability of funds
**Key Questions DARPA Asks**: ### NSTC Review Criteria
**Core Evaluation Dimensions**:
1. **Innovation (創新性)**: Novelty of concept and approach.
2. **Feasibility (可行性)**: Methodology rigor and preliminary data.
3. **PI Capability (主持人能力)**: Track record and expertise.
4. **Value (價值)**: Academic contribution and societal/industrial impact.
For detailed review criteria by agency, refer to `references/review_criteria.md` and `references/nstc_guidelines.md`.
- **What if you succeed?** (Impact if the research works) - **What if you succeed?** (Impact if the research works)
- **What if you're right?** (Implications of your hypothesis) - **What if you're right?** (Implications of your hypothesis)
- **Who cares?** (Why it matters for national security) - **Who cares?** (Why it matters for national security)

733
scientific-skills/research-grants/references/nstc_guidelines.md Normal file
View File

@@ -0,0 +1,733 @@
# Taiwan NSTC (National Science and Technology Council) Proposal Guidelines
> ⚠️ **IMPORTANT DISCLAIMER**: This guide is based on publicly available information and general academic writing principles. **Always consult the official NSTC website and your specific program's solicitation for the most accurate and up-to-date requirements.** Requirements may vary by field, program type, and year.
## Overview
**Official Name**: 國家科學及技術委員會 (National Science and Technology Council, NSTC)
**Former Name**: 科技部 (Ministry of Science and Technology, MOST)
**Official Website**: https://www.nstc.gov.tw/
**Mission**: Advance Taiwan's scientific and technological development through research funding, with emphasis on scientific breakthrough, industrial application, and societal impact.
---
## CM03: Research Proposal Content (研究計畫內容)
CM03 is the core technical document of your NSTC proposal. It is officially titled "Contents of Grant Proposal" (計畫書本文).
### Official Format Requirements
Based on official NSTC documentation:
**Paper Size**: A4 (29.7 cm × 21 cm)
**Font**:
- Chinese: PMingLiU (新細明體) or BiauKai (標楷體)
- English: Times New Roman or Arial
- Size: 12-point minimum
**Spacing**: Single space for English; no extra spacing between lines for Chinese
**Page Limits** (varies by field and program type):
- **Humanities**: Individual 1-year: 30 pages; Multi-year: 45 pages
- **Engineering**: Individual 1-year: 20 pages; Multi-year: 25 pages
- **Natural Sciences**: Individual: 30 pages; Integrated: 45 pages
- **Life Sciences**: Individual: 25 pages
- **⚠️ CRITICAL**: Page limits include references and figures. Exceeding limits may result in automatic rejection.
**File Format**: PDF recommended for submission
---
## Required Content Sections
Based on official CM03 templates, the proposal must include:
### 1. Abstract (摘要)
**Requirements**:
- **Chinese abstract**: Maximum 500 characters
- **English abstract**: Maximum 500 words
- **Keywords**: 3-5 keywords in both languages
**Content**:
- Research background and problem statement
- Research objectives
- Key methods and approaches
- Expected outcomes and impact
### 2. Research Background and Objectives (研究計畫之背景及目的)
**Required Elements**:
- Problem statement and significance
- Research originality and innovation
- Expected impact
- Review of domestic and international related research
- Important references with critical evaluation
- **For continuing projects**: Progress from previous year
### 3. Research Methods, Steps, and Timeline (研究方法、進行步驟及執行進度)
**Required Elements**:
- Research principles and methodology
- Justification for chosen methods
- Innovative aspects of the approach
- Anticipated problems and solutions
- Equipment and instrumentation needs
- **For international travel**: Justification and expected benefits
- **Timeline**: Year-by-year breakdown of activities
### 4. Expected Outcomes (預期完成之工作項目及成果)
**Required Elements**:
- Expected research tasks (by year)
- Personnel training plans
- Expected outputs:
- Journal articles (specify target journals)
- Conference papers
- Patents
- Technology transfer
- Other deliverables
---
## 114年度 (2025) Application Requirements
Based on official announcements:
**Application Method**: Fully online through NSTC Academic Research Service Network (學術研發服務網)
**Project Start Date**: Most projects begin August 1, 2025 (114年8月1日)
**Academic Ethics Requirement**:
- First-time applicants and first-time participants must complete **at least 6 hours** of academic ethics training within 3 years before submission
- Must provide certification
**Thesis Disclosure**:
- If proposal content involves student theses supervised by the PI, it must be clearly disclosed or cited
- Already published work (including student theses) should not be hidden as new research content
---
## Budget Categories (經費編列)
Based on official guidelines:
**Personnel (人事費)**:
- Postdoctoral researchers
- Research assistants
- Part-time staff
- **Note**: PI salary typically not allowed
**Equipment (設備費)**:
- Items exceeding NT$10,000 with service life > 2 years
- Items exceeding NT$200,000 may require price appraisal
**Consumables (耗材費)**:
- Lab supplies, reagents, software licenses
**Travel (差旅費)**:
- Domestic and international conferences
- Research collaborations
**Other (其他費用)**:
- Publication fees, data collection, outsourcing
---
## Review Criteria
**Note**: Specific scoring weights are not publicly disclosed by NSTC. The following are general evaluation dimensions based on academic practice:
1. **Innovation (創新性)**: Novelty of concept and approach
2. **Feasibility (可行性)**: Methodology soundness and preliminary data
3. **PI Capability (主持人研究能力)**: Track record and expertise
4. **Value (價值)**: Academic contribution and societal/industrial impact
---
## Official Resources
**NSTC Website**: https://www.nstc.gov.tw/
**Application System**: Access through "學術研發服務網" (Academic Research Service Network)
**Help Desk**:
- Computer/System Issues: 0800-212-058 or (02)2737-7592
- Regulation Questions: (02)2737-7440, 7568, 7847, 7980, 8010
**Important**: Always download the latest application forms and guidelines from the official NSTC website under "專題研究計畫專區" (Research Project Area).
### LaTeX Templates
For those who prefer LaTeX for proposal writing, there are excellent community-contributed templates available:
#### Official CTAN Package (Recommended)
**nstc-proposal** - Professional LaTeX classes for NSTC proposals:
- **GitHub**: https://github.com/L-TChen/nstc-proposal
- **CTAN**: Available via `tlmgr install nstc-proposal`
- **Supports**: Both CM03 and CM302 (bibliography format)
- **Features**:
- Compatible with pdfLaTeX and XeTeX
- Bilingual support (Chinese/English)
- Pre-defined section commands (`\ProposalBackground`, `\ProposalMethod`, `\ProposalPlan`, `\ProposalIntegration`)
- Multiple font options (standard, Libertine, KaiTi)
- Proper formatting for NSTC requirements
**Installation**:
```bash
# Via TeX package manager (easiest)
tlmgr install nstc-proposal
# Or manual installation from GitHub
git clone https://github.com/L-TChen/nstc-proposal.git
cd nstc-proposal
latex nstc-proposal.ins
```
**Basic Usage Example**:
```latex
\documentclass{nstc-cm03}
\usepackage{microtype}
\begin{document}
\ProposalBackground
% Your content here
\ProposalMethod
% Your content here
\ProposalPlan
% Your content here
\nocite{*}
\bibliographystyle{plain}
\bibliography{references}
\end{document}
```
#### Alternative Templates
**Engineering Division Template**:
- GitHub: https://github.com/mcps5601/NSTC-proposal-LaTeX
- Provides CM03 format specifically for Engineering Division (工程司)
- **Note**: Format requirements may differ by division
**Overleaf Templates**:
1. **audachang's CM03 Template** (Recommended for Overleaf users):
- GitHub: https://github.com/audachang/taiwan-nstc-cm03-template
- Overleaf: Direct import from GitHub
- **Features**:
- Includes official CM03.doc file for reference
- Uses XeCJK with BiauKai (標楷體) font for Traditional Chinese
- Organized structure with separate section files (`background.tex`, `methods.tex`, `expected_outcomes.tex`)
- **Important**: Must use XeLaTeX or LuaLaTeX compiler
- Based on Chen Wen-sheng's template
2. **Other Overleaf Templates**:
- Search for "國科會研究計畫內容: CM03" on Overleaf
- Various community-contributed templates available
> ⚠️ **Important**: These are community-contributed templates. Always verify that the format complies with the latest official NSTC requirements for your specific field and program type. The `nstc-proposal` CTAN package is regularly maintained and is the most reliable option.
---
## Practical Insights from Reviewers
> 📚 **Source**: This section is based on "國科會計畫撰寫經驗分享" by Prof. Huang You-Ping (黃有評), President of National Penghu University of Science and Technology. These insights reflect the **reviewer's perspective** and are particularly relevant for Engineering Division proposals.
> ⚠️ **Important**: Scoring thresholds and specific criteria may vary by division (Humanities, Engineering, Natural Sciences, Life Sciences, etc.). Always check with your specific field's requirements.
### Understanding the Scoring System
Based on Engineering Division (工程司) - Automation/Control field experience:
**Scoring Thresholds**:
- **92+ points (Top 5%)**: Outstanding research level - eligible for Distinguished Research Award (傑出研究獎)
- **88+ points (Top 15%)**: Required threshold if applying for a second concurrent project
- **81+ points (Top 54-55%)**: **Passing threshold** - proposals scoring 81 or above are recommended for approval
- **80 points or below**: Not recommended for approval
**Key Insight**: The difference between "passing" (81) and "excellent" (88+) often lies in the strength of preliminary data, clarity of innovation, and demonstrated feasibility.
---
### Section-by-Section Writing Strategies
#### Abstract (摘要)
**Reviewer Expectations**:
- Must demonstrate **innovation** and **problem-solving strategy** immediately
- Should capture attention in the first reading
- Clearly state what makes this proposal different from existing work
**Critical Question**: Does the abstract make the reviewer want to read more?
#### Research Background and Motivation (研究背景及目的)
**What Reviewers Look For**:
- **Clear problem definition**: Is the core problem well-defined?
- **Reasonable design and objectives**: Are the goals achievable and well-justified?
- **Logical flow**: Does the background naturally lead to your research objectives?
**Common Weakness**: Vague problem statements that don't clearly identify what gap you're filling.
#### Literature Review (文獻探討)
**Quality Over Quantity**:
- Select **highly relevant** literature, not just many papers
- **Critical synthesis**: Don't just list papers - analyze strengths, weaknesses, and gaps
- **Recency matters**: Include publications from the **last 2-3 years** to show awareness of current state-of-the-art
- **Strategic positioning**: Use literature review to guide readers toward your research objectives
**Reviewer's Perspective**: A well-curated 20-paper review with critical analysis is far superior to a 50-paper list without synthesis.
#### Research Methods and Implementation (研究方法、進行步驟及執行進度)
**Feasibility is Critical**:
- **Avoid over-idealization**: Proposals that are too ambitious without clear mitigation strategies often fail
- **Logical progression**: Each step should follow naturally from the previous one
- **Comparison with existing methods**: Clearly show how your approach differs and why it's better
- **Contingency planning**: Address potential problems and provide alternative approaches
**Red Flags for Reviewers**:
- Methods that are too difficult without demonstrated capability
- Lack of logical connection between steps
- No discussion of potential challenges
- Missing preliminary data for novel approaches
#### Expected Outcomes (預期完成之工作項目及成果)
**Be Specific and Quantifiable**:
- ✅ **Good**: "Improve system efficiency by 15% compared to baseline method X"
- ❌ **Weak**: "Improve system efficiency"
**Include Multiple Dimensions**:
- **Academic value**: Target journals and expected number of publications
- **Economic benefits**: Potential industrial applications
- **Talent cultivation**: Number and level of students to be trained
---
### Budget Preparation Tips
**Alignment with Research Plan**:
- Every budget item should directly support a specific research activity
- Personnel costs should reflect actual time commitment
- Equipment justification should explain why existing facilities are insufficient
**International Conference Travel**:
- Typical budget: NT$70,000 - 100,000
- **Must justify**: Explain your track record of international conference participation and contributions
- Show how conference attendance benefits the research
**Reviewer's Check**: Does the budget match the proposed activities? Are there unexplained large expenses?
---
### Strategic Career Advice
**For New Faculty**:
1. **Always apply**: New investigators have certain advantages - don't miss the opportunity
2. **Build foundation**: Use undergraduate research projects (大專學生研究計畫) to develop preliminary data
3. **Self-assessment**: Use the review criteria checklist to evaluate your proposal before submission
**Building Academic Visibility**:
- Join professional societies (e.g., IEEE, CAA)
- Serve as reviewer for journals and conferences
- Take on roles as Associate Editor (AE) or board member
- **Why it matters**: Reviewers are more likely to recognize and trust researchers who are active in the community
---
### Preparation and Mindset
**Timeline**:
- **Start early**: Successful proposals require multiple revisions
- **Iterate**: Don't wait until the deadline to start writing
- **Seek feedback**: Have colleagues review your draft
**Handling Rejection**:
- **Learn from feedback**: Carefully review all reviewer comments
- **Revise and resubmit**: Address criticisms in next submission
- **Consider alternatives**: If fundamental issues exist, consider different program types or focus areas
**Professional Presentation**:
- **Figures and tables**: Must be clear, numbered, and properly labeled
- **Formatting**: Professional layout demonstrates attention to detail
- **Proofreading**: Typos and formatting errors suggest carelessness
---
### Self-Assessment Checklist
Before submitting, ask yourself:
**Innovation**:
- [ ] Is my approach genuinely novel or just incremental?
- [ ] Have I clearly explained what's new compared to existing work?
- [ ] Do I have evidence (preliminary data) that my innovation is feasible?
**Feasibility**:
- [ ] Are my methods well-described and logical?
- [ ] Do I have the necessary expertise and resources?
- [ ] Have I addressed potential problems?
- [ ] Is my timeline realistic?
**Impact**:
- [ ] Are my expected outcomes specific and measurable?
- [ ] Have I explained both academic and practical value?
- [ ] Does my proposal align with national priorities or industrial needs?
**Presentation**:
- [ ] Are all figures clear and properly labeled?
- [ ] Is the writing clear and free of errors?
- [ ] Does the budget align with proposed activities?
- [ ] Have I included all required sections?
---
## Advanced Writing Strategies from Government Reviewers
> 📚 **Sources**: This section integrates insights from two comprehensive guides:
> 1. "如何提升政府科技發展計畫書撰寫品質" by **Prof. Guo Yao-Huang (郭耀煌教授)**
> 2. "如何提升政府科技發展計畫書撰寫品質" by **President Wei Yao-Hui (魏耀揮校長)**, Mackay Medical College
>
> These guides are based on extensive experience reviewing government science and technology proposals (including NSTC and other ministry programs).
### The Closed-Loop Logic Framework
**Core Principle**: A high-quality proposal must demonstrate complete logical coherence from problem to performance.
**The Loop**:
```
Problem Discovery → Goal Definition → Strategy Formulation →
Concrete Measures → Execution Plan → Performance Indicators (KPI)
```
**Critical Requirement**: Every element must connect logically.
**Example of Broken Logic**:
- ❌ **Goal**: Improve industrial technology
- ❌ **Strategy**: Provide student scholarships
- **Problem**: The strategy doesn't directly support the goal
**Example of Closed Logic**:
- ✅ **Goal**: Improve industrial technology
- ✅ **Strategy**: Develop advanced manufacturing process
- ✅ **Measures**: Establish testing facility, train engineers
- ✅ **KPI**: Achieve 15% efficiency improvement, train 20 engineers
---
### SMART Principle for Proposal Planning
Before writing, ensure your proposal meets **SMART** criteria:
| Criterion | Meaning | Application |
|-----------|---------|-------------|
| **S**pecific | Concrete goals | Define exact technical metrics (e.g., "improve accuracy to 95%") |
| **M**easurable | Quantifiable KPIs | Use numbers, percentages, counts |
| **A**chievable | Realistic scope | Match available resources, personnel, equipment, budget |
| **R**ealistic | Scientific basis | Grounded in data and logical reasoning |
| **T**imely | Clear timeline | Specific milestones with dates |
---
### Four Dimensions of Review Criteria
Reviewers evaluate proposals across four key dimensions:
#### 1. **Necessity (需求性)**
- Does it align with national science and technology policies?
- Is there urgent need for this research?
- Why must this problem be solved **now**?
- Why is **your institution** the right one to do this?
**Weak Proposal**: Generic problem statement without urgency
**Strong Proposal**: Cites specific policy documents, demonstrates time-sensitive need
#### 2. **Feasibility (可行性)**
- Are the goals achievable within the proposed timeline?
- Is the team qualified (track record, expertise)?
- Are the methods sound and well-justified?
- Is the management plan realistic?
**Red Flag**: Overly ambitious goals without preliminary data or risk mitigation
#### 3. **Appropriateness (適當性)**
- Does the budget match the work scope?
- Are personnel allocations reasonable?
- Is existing equipment utilized effectively?
- Are expensive items properly justified?
**Reviewer's Question**: Why do you need this expensive equipment when similar facilities exist?
#### 4. **Impact and Benefits (效益與影響)**
- Beyond academic output, what are the societal effects?
- Economic benefits or industrial applications?
- Environmental, health, or national security impacts?
- Long-term sustainability?
**Key Insight**: Reviewers increasingly value **societal impact** over pure academic metrics.
---
### Performance Indicators (KPI): The Three Levels
Understanding the difference between input, output, and outcome is critical:
| Level | Type | Examples | Reviewer Value |
|-------|------|----------|----------------|
| **Input** | Resources invested | Personnel, budget, equipment | Basic requirement |
| **Output** | Direct products | Papers, patents, conferences | Minimum expectation |
| **Outcome** | Real-world impact | Industry adoption, health improvement, policy influence | **High value** |
**Example Comparison**:
- ❌ **Weak KPI**: "Publish 3 papers" (output only)
- ✅ **Strong KPI**: "Publish 3 papers in Q1 journals AND transfer technology to 2 companies, generating NT$5M in licensing revenue" (output + outcome)
**KPI Best Practices**:
- **Relevance**: Directly tied to project goals
- **Ease**: Simple to measure and verify
- **Credibility**: Based on realistic projections
- **Cost-efficiency**: Achievable within budget
**Progressive Targets**: Show year-by-year progress, not just final goals
- Year 1: 30% completion
- Year 2: 70% completion
- Year 3: 100% completion + sustainability plan
---
### Practical Analysis Tools
#### SWOT Analysis
Use SWOT to position your proposal strategically:
| Strengths | Weaknesses |
|-----------|------------|
| Your unique expertise | Resource limitations |
| Existing facilities | Lack of certain skills |
| Strong track record | Time constraints |
| Opportunities | Threats |
|---------------|---------|
| Policy alignment | Competing teams |
| Industry partnerships | Technology changes |
| Emerging trends | Funding cuts |
**Critical**: Don't just list SWOT - **provide response strategies** for Weaknesses and Threats.
**Example**:
- **Weakness**: Lack of high-performance computing cluster
- **Response**: Partner with National Center for High-performance Computing (國網中心)
#### Fishbone Diagram (魚骨圖)
Use fishbone diagrams to demonstrate deep problem understanding:
```
Main Problem
↑
┌───────┬────────┼────────┬───────┐
Factor 1 Factor 2 Factor 3 Factor 4
│ │ │ │
Sub-causes Sub-causes Sub-causes Sub-causes
```
**Purpose**: Show reviewers you've thoroughly analyzed root causes, not just symptoms.
#### Gantt Chart
For complex multi-year projects, include Gantt charts to show:
- Task dependencies
- Resource allocation over time
- Milestones and deliverables
- Risk management checkpoints
**Professional Presentation**: Use visual tools to demonstrate project management capability.
---
### Budget Preparation: Critical Details
#### Necessity and Reasonableness
**The Two Questions Every Budget Item Must Answer**:
1. **Why is this necessary?** (Link to specific research activity)
2. **How was this calculated?** (Show detailed breakdown)
**Example - Equipment Justification**:
- ❌ **Weak**: "High-performance workstation: NT$150,000"
- ✅ **Strong**: "High-performance workstation (Intel Xeon 32-core, 128GB RAM, RTX 4090 GPU) for deep learning model training: NT$150,000. Current lab computers (8GB RAM) cannot handle the 50GB dataset required for Aim 2. Estimated training time reduction from 2 weeks to 2 days."
#### Budget Category Separation
**Critical Rule**: Strictly separate "recurrent" (經常門) and "capital" (資本門) expenses.
**Recurrent (經常門)**:
- Personnel salaries
- Travel expenses
- Consumables
- Publication fees
**Capital (資本門)**:
- Equipment ≥ NT$10,000 with lifespan ≥ 2 years
- Items ≥ NT$200,000 may require price comparison
**Forbidden**: Using science and technology funds for general administrative work
#### Outsourcing (委辦費用)
If including outsourcing costs:
- Specify exact scope of work
- Explain why in-house execution is not feasible
- Describe selection and oversight procedures
- Provide cost breakdown
#### International Conference Travel
**Typical Range**: NT$70,000 - 100,000
**Required Justification**:
- Your track record of international presentations
- Specific conference name and dates (if known)
- How attendance benefits the research (networking, collaboration, dissemination)
- Why this conference is important for your field
---
### Common Review Comments to Avoid
Based on actual reviewer feedback from government proposals:
#### 1. **Vague Objectives**
- ❌ "Promote development of..."
- ❌ "Research on..."
- ✅ "Develop algorithm achieving 95% accuracy on benchmark X"
#### 2. **Redundancy and Overlap**
- **Problem**: Multiple agencies funding similar work
- **Solution**: Clearly differentiate from existing programs; coordinate with other ministries before submission
#### 3. **Lack of Continuity Explanation**
- **For continuing projects**: Must explain relationship between previous results and new proposal
- Show how you're building on (not repeating) past work
#### 4. **Technology Push Without Market Pull**
- **Problem**: Developing technology without considering industry needs or market readiness
- **Solution**: Include industry partner letters, market analysis, or user needs assessment
#### 5. **Ignoring Negative Impacts**
- **Common oversight**: Privacy concerns, environmental impact, ethical issues
- **Solution**: Include risk assessment and mitigation strategies
#### 6. **Excessive Administrative Overhead**
- **Problem**: Too many project management offices (PMO) or coordinators
- **Solution**: Justify administrative structure based on project complexity
#### 7. **Missing Customer Definition**
- **Question**: Who will use your research results?
- **Answer**: Clearly define your target users/beneficiaries
---
### Writing for the Reviewer
**Remember**: You're writing for busy reviewers, not for yourself.
**Best Practices**:
1. **Use visual aids**: Replace dense text with figures, tables, flowcharts
2. **Data-driven**: Support claims with specific numbers and citations
3. **Objective correctness**: Verify all data and calculations
4. **Logical flow**: Each section should naturally lead to the next
5. **Professional polish**: Clean formatting, no typos, consistent terminology
**Critical Question**: After reading your abstract, does the reviewer **want** to read more?
---
### Policy Alignment
**Essential**: Connect your research to national priorities.
**How to Demonstrate Alignment**:
- Cite specific government policy documents (e.g., "六大核心戰略產業")
- Reference national development plans
- Show how your research addresses societal needs
- Link to ministry-specific priorities
**Example**:
"This research directly supports Taiwan's '5+2 Innovative Industries' initiative, specifically the biomedical sector, by developing..."
---
### Exit Strategy (For Multi-Year Projects)
**Requirement**: Long-term projects must include sustainability plans.
**Key Questions**:
- What happens when funding ends?
- How will results be maintained or transferred?
- What are the success/failure criteria for early termination?
**Components**:
- Technology transfer plan
- Industry partnership agreements
- Follow-on funding strategy
- Publication and dissemination plan
---
### Evaluation Mechanisms
**For public service projects**: Include feedback and assessment systems.
**Components**:
- User satisfaction surveys
- Performance metrics tracking
- Regular review milestones
- Adjustment mechanisms based on feedback
---
## Common Mistakes to Avoid
1. **Exceeding page limits** → Automatic rejection
2. **Missing required sections** → Incomplete application
3. **Incorrect font or formatting** → Non-compliance
4. **Lack of preliminary data** (for applicable programs) → Reduced competitiveness
5. **Vague methodology** → Feasibility concerns
6. **No connection to Taiwan context** → Lower impact score
---
## Final Checklist
Before submission:
- [ ] Check specific program solicitation for field-specific requirements
- [ ] Verify page limit for your field and program type
- [ ] Complete academic ethics training (if required)
- [ ] Prepare both Chinese and English abstracts
- [ ] Include all required forms (CM01, CM02, CM03, etc.)
- [ ] Verify all formatting requirements
- [ ] Proofread for errors
- [ ] Submit through official online system before deadline
---
## Disclaimer
**This guide is for reference only.** Official requirements may change annually and vary by program. **Always consult**:
1. The latest official NSTC announcements (徵求公告)
2. Your specific program's application guidelines
3. Your institution's research office
4. Senior colleagues in your field
For the most authoritative information, visit: **https://www.nstc.gov.tw/**