Your Unofficial Guide to Reviewing Papers
This guide is an unofficial guide for academic paper reviewing, professional conduct, and academic writing quality at TU Delft.
📋 Table of Contents
Professional Conduct
Document Quality Standards
Spelling and Grammar:
- Always spell-check your documents before distribution
- Use tools like
ispell
,detex
,spell
,latex2html
, WinEdit, or Word - Asking readers to fix trivial errors is discourteous
Document Structure:
- Page Numbers: Always include page numbers in documents and presentations
- Table of Contents: Essential for documents longer than a few pages
- Additional Lists: For large documents (theses), include:
- List of figures
- Terminology
- Abbreviations
- Symbols
- Definitions
- Theorems
- Index (very helpful for long documents)
Collaboration and Feedback
Paper Submission Process:
- Give co-authors sufficient time to review papers
- Update git repository with submitted PDF in snapshot directory
- Repeat process for final accepted version
Plagiarism Prevention:
- Clearly indicate when reusing text, figures, or data from other documents
- Avoid reusing text from your own papers in new papers (except extended journal versions)
- Self-plagiarism is not acceptable and leads to paper rejection
Paper Structure
Standard Academic Paper Structure
1. Title
2. Abstract
3. Introduction
4. Related Work
5. Problem Definition
6. Solution/Approach
7. Experiments/Evaluation
8. Results
9. Discussion
10. Conclusions
11. References
Key Requirements:
- Follow this structure for acceptance
- Reviewers expect this organization
- Deviations may lead to rejection
Detailed Paper Structure Guidelines
Title
- Nano version of the paper: What is it about?
Abstract
- Micro version of the paper, focusing on:
- Problem
- Novelty
- Claims
- Possibly some high-level results
- Content:
- Trends (leading to the problem)
- Problem statement
- Novelty & claims (in terms of approach, solution, results)
- [Optional] High-level results
Introduction
- Mini version of the paper, focusing on:
- Problem
- Novelty
- Claims
- High-level approach & solution
- Content:
- Trends (leading to the problem)
- Problem statement
- Novelty & claims
- Approach & solution
- Structure of paper
- Note:
- Repetition and recursion are useful
- Explain the structure of your paper to the reader
(e.g., at the end of the introduction, and by linking sections)
Related Work (or before Conclusions)
- Explain how your work:
- Differs from,
- Improves on,
- Extends,
other work
- Not just a list of other works!
(High-level Overview of) Approach & Solution
- This section can be skipped, especially in shorter papers.
Your Novel Theory / Experimental Setup / Architecture / …
- Short introduction, linking to the previous section
- Subsections
- Short conclusion, linking to the next section
Experimental Results
- Explain the reason for the experiments
(Hint: to prove your claims) - Then describe them
- Explain (not just describe) your experimental results
- Show that they prove your claims
Conclusions
- State that you solved the problem with your solution
- State that you proved your claims
- Include:
- (Reminder of) Problem statement
- (Reminder of) Solution (approach)
- Results proving the claims
Document Reviewing
Three-Level Review Process
Level 1: 3-5 Minutes (Fail-Fast)
- Read only: title, abstract, conclusions, references
- Check: good English, clear problem, relevance, solution, citations
- Reject if any criteria not met
- Most people read papers this way
Level 2: 10-15 Minutes (High-Level)
- Also examine: figures, high-level approach, results section
- Check: solution understanding, experiment validity, claims proven
- Reject if criteria not met
Level 3: 1+ Hour (Detailed)
- Read everything, understand everything
- Almost no-one does this
- Reject only for real faults
Review Checklist
Content Verification:
- Do title, abstract, introduction, and conclusions match?
- Are claims in abstract & introduction justified?
- Is problem defined well?
- Is solution and concepts defined well?
- Are limitations and constraints defined clearly?
Structure and Presentation:
- Is paper well-structured with proper sections, layout, figures?
- Is English understandable?
- Is there a related work section?
- Is related work correct and fair to other papers?
Experiments and Results:
- Are experiments defined in sufficient detail?
- Are experiments presented well (understandable graphs)?
- Are conclusions from experiments justified?
- What is the novelty of the paper?
Review Writing Guidelines
Professional Conduct:
- Be polite and don't offend authors
- Write impersonal reviews: "the text is difficult to understand" not "I don't understand"
- Give suggestions for improvement
- Review to same standard regardless of venue
Revision Responses:
- Include cover letter thanking reviewers and editor
- Respond to each comment in detail
- Ensure each response results in visible changes
Understanding Professor Feedback
Common Symbols and Abbreviations
Approval Indicators:
- C: OK
- ~: OK-ish
- ✓: Good
Problem Indicators:
- ?: Unclear (larger = bigger problem)
- ×, n: Not OK (larger = bigger problem)
- ≠: Inconsistent, incorrect
Structural Changes:
- large ~ between words: Swap these parts
- \: Break into two paragraphs or lines
- ×: Join two parts (paragraphs, sentences)
Logical Connectives:
- ⇒: Implies
- /⇒: Non sequitur (does not follow)
- ∴: Therefore
- ∵: Because
References:
- ②, ①, ②: Text referred to later/elsewhere
Common Abbreviations:
- o/w: Otherwise
- w/o: Without
- w/: With
- p/pp: Page/pages
- def/(un)def'd: Definition/(un)defined
- emph/it: Emphasize/italicize
- ip/op: Input/output
- (dis)cts: (Dis)continuous
- reqˢ/respˢ: Request(s)/response(s)
- trans: Transaction
- beh: Behaviour
- eqn: Equation