This questionnaire helps researchers structure and plan their scientific papers through guided
reflection questions.
Based on Michael Black's
writing guidance,
this tool walks you through the process of:
Part 1: Idea Formulation - Clarify your research idea, core contribution, and
narrative
Part 2: Content Preparation - Plan each section of your paper methodically
Part 3: Best Practices - Consider the reviewer perspective and anticipate questions
Your answers are automatically saved in your browser. You can export your work as PDF, Markdown, or
JSON,
and import previously saved work. You can also manage multiple questionnaires for different papers.
This section focuses on clarifying your research idea, identifying your core contribution, and mapping out
your narrative.
1. Defining Your Core Idea
What is the central problem you are trying to solve? (Be specific and concise)
Why is this problem important, and why should anyone care? (Focus on impact and
relevance)
What specific impediment has prevented this from being solved effectively by
previous work? (Identify the bottleneck or limitation)
What is your hypothesis? Is it clearly testable, and how will you verify its
truth or falsity? (Must be a falsifiable statement)
What is your "Nugget"? (Your single most important key insight that makes the
impossible, possible. This is the "how you see the world differently" aspect.)
2. Understanding Your Audience and Impact
Who is your primary audience for this paper? (e.g., specific sub-field, general
AI/ML researchers, practitioners)
Who will likely build upon your work? (Consider the future impact and potential
follow-up research)
3. Crafting Your Narrative (Goal, Problem, Solution)
What is the "Goal" you will define for the reader, and why is it desirable? (What
do you want the reader to desire?)
What is the "Problem" you will present that prevents this goal from being
achieved? (What creates tension/drama?)
What is your "Solution" (your clever idea/Nugget) that overcomes this obstacle?
(How do you become the hero?)
Can you outline a layered Goal → Problem → Solution cycle if your paper has
multiple insights?
4. Distilling Your Message
What is your elevator pitch? (Describe your paper's core idea in three sentences
or fewer)
What is your "teaser" image/figure? (What single visual would explain your core
idea to someone instantly?)
5. Related Work and Evaluation
What are the key previous works directly relevant to your research?
Specifically, where do these previous works fall short, creating a gap for your
research? (This illuminates your path forward)
How will you quantitatively evaluate your method? (What metrics, datasets,
comparisons?)
What is your "demo"? How will you convincingly show that your idea works? (Beyond
quantitative metrics, what qualitative demonstration?)
Part 2: Section-by-Section Content Preparation
This section guides you through preparing the specific content for each part of your paper.
1. Title and Acronym
Brainstorm 3-5 potential titles for your paper. (Aim for short, suggestive
poetry)
What relevant nouns and verbs describe your method? (For acronym generation)
Based on the above, brainstorm 3-5 potential acronyms. (Aim for catchy,
pronounceable, semantic, and searchable)
2. Abstract
Using the "Mad Libs" template as a guide, fill in the blanks with your specific content.
[Your topic] is widely used in computer vision/AI/etc. and has applications in
[your application].
Recent work has addressed this problem by [describe previous approach].
Unfortunately, all of these approaches [describe their key limitation].
In contrast, we [insert your nugget/key insight].
This fixes [the previous problem], however, it does not solve [a new, smaller
problem].
Consequently, we develop a novel [your technical contribution].
While promising, [describe a challenge with your solution] is non-trivial.
Therefore, we further [describe your final technical piece].
We evaluate [your method] qualitatively and quantitatively on [your datasets] and
find that it is more accurate than the state of the art.
3. Introduction
How will you state why your topic (X) is important and why it remains unsolved?
(Focus on the problem, not your interest)
What are your explicit contributions that you will state in the introduction?
(e.g., "Our contributions are...")
How will you introduce your "Nugget" and frame it as the solution to the core
problem?
4. Previous / Related Work
What are the main themes or common approaches that group the previous papers
you've identified?
How will you structure this section to provide insight into the field's history,
leading up to your work, rather than just a list of citations?
5. Experiments / Results / Evaluation
What are the specific quantitative metrics you will use to evaluate your method?
What datasets will you use, and why are they appropriate?
What baseline methods or state-of-the-art approaches will you compare against?
How will you present your qualitative results or "demo" to convincingly show your
idea works? (e.g., specific examples, visualizations)
What are the key findings or takeaways from your experiments that you want to
highlight?
6. Discussion / Limitations / Future Work
What are the clear limitations or weaknesses of your current work? (Be honest and
objective)
How will you frame these limitations as clear opportunities for future research?
What broader implications or interesting discussions can arise from your results?
7. Conclusion
How will you succinctly summarize the main contribution and findings of your
paper?
What is the single most important message you want the reader to take away from
your paper?
Part 3: General Best Practices & Strategy
This section helps you consider the overall approach and reviewer's perspective.
1. Thinking Like a Reviewer
What do you anticipate a reviewer will identify as the key ideas and significance
of your work?
What do you expect a reviewer to highlight as the strengths of your paper?
What potential weaknesses or unsupported claims might a reviewer question? (How
will you address or mitigate these?)
2. Readiness and Risks
Do you have all the necessary data to support your claims and evaluations?
What are the key risks to completing this paper on time or with the desired
results? (e.g., missing data, unresolved bugs, lack of specific expertise)