Lec12 Organize Paper
正文
Form follows Venue
Our writing form is going to be based on where we intend to publish our paper.
Examples
| CoCa | ||
|---|---|---|
| Recent preprint by Microsoft (at the time of this lecture). Formatted like an ICLR conference submission for 2023. | Recent publication at CVPR 2022 by Facebook AI Research. | Recent publication at Transactions of Machine Learning Research by Google Research. |
Sequence of the Sections
| CoCa | ||
|---|---|---|
| Abstract Introduction Methods Experiments Related Work Conclusion | Abstract Introduction Background FLAVA: A Foundational Language And Vision Alignment Model Experiments Conclusion | Abstract Introduction Related Work Approach Experiments Broader Impacts Conclusion |
Some commonalities:
- 6-7 Sections (inclusive of abstract).
- An abstract, introduction to start the paper, and conclusion to end it.
- Related Work / Background is either the section after the introduction or before the conclusion.
- The Methods or Approach section describing the approach is next.
- Experiments come after the methods section.
Sequence of the Figures
| Method overview diagram Comparison to other models on 2 tasks Comparison to other models on 2 other tasks Comparison to other models on 1 more task Comparison to ablations | Method overview diagram Comparison of capabilities recent models in different modalities Lower level overview of model Representative examples from various subsets of our pretraining dataset Datasets used for pre-training Comparison to ablations Comparing to different eval settings Comparing to previous models Performance difference to one model on different tasks | Overview of the method Illustration of the architecture and objectives Ablation analysis of size Illustration of method for video recognition Comparison of model performance with other models across tasks (x 2) Image classification scaling performance of model sizes. Comparison of model performances on some tasks (x5) Curated samples of input and output predictions Comparison to ablations |
Some commonalities:
- 5-15 figures
- Starts with a method overview diagram. Sometimes has more illustrations of the method for different tasks, or a lower level illustration.
- Shows results comparing models across different tasks, with a table/figure for different sets of tasks. These comparisons are typically to previous models.
- Shows results comparing method ablations.
- Sometimes shows examples of the input and output predictions.
Local Structure
Definition of Local Structure: we’re going to look at each of the individual sections and break down their structure – we’re going to call this the local structure.
Structure of the Abstract
Let’s start with the abstract – I want us to go through the abstract, line by line, and write down the purpose of each sentence. In particular, we are going to write down the question each sentence answers.
| What is the solution introduced in the paper? What is the key idea of the solution at a high level? What are the key components of the solution (x2)? What are the strengths of the solution? What are notable results (x2)? | What is the background for the class of models? What is the key gap with previous models? What is the key desiderata of a solution? What is the solution introduced in the paper and what are notable takeaways? | What is the significance of the research topic? What is the solution introduced in the paper? What are the key components of the solution that are different from previous approaches? What are the components of the solution? What are the strengths of the solution? What are notable results, mentioning tasks and numbers? |
Notes:
- 116 words, 110 words and 254 words respectively.
- Same 5-6 components, but no order except end with notable results/takeaways.
Structure of the Introduction
We’re going to look at the introduction paragraph-by-paragraph. For each paragraph, let’s write down the question that each sentence answers.
| How successful has the class of approaches under consideration previously been? How do previous approaches in the class tackle the problem? What is the solution introduced in the paper? What are the key components of the solution? What are the strengths of the solution? What are the experiments that are performed? What do the experimental results demonstrate? What are the main contributions, bulletted? What is the key idea of the solution? What is the strength of the solution? What do the results positively indicate? | How successful has the class of approaches under consideration previously been? Can you give a couple of examples? What is wrong with a class of previous approaches (x3)? What is another possible approach? What is wrong with that class of approach? What is the key desiderata of a solution? What is the key desiderata of a solution? What is the solution introduced in the paper? What are the key components of the solution? What do the experimental results positively demonstrate? What are the strengths of the solution? | How successful has the class of approaches under consideration previously been? What is their key disadvantage? What is an approach to the problem? Why is it problematic? What is a solution to that problem or an alternative approach? What is wrong with that? What is the solution introduced in the paper? What are the key components of the solution (x 6) ? What is the strength of the solution? What do the experimental results positively demonstrate? |
Notes:
- 4-6 paragraphs.
- Very similar starts and ends!
- Includes how the previous class of approaches tackle the problem, and what’s wrong with them.
- Includes what the main components of the solution and its strengths are.
Structure of the Related work
We now repeat the exercise with the related work. This time, for each paragraph, we are going to write down the purpose of that paragraph. You could do this exercise at a sentence level too, but it is typically less structured than an intro, so we do this at the paragraph-level.
| Overall related work approach description, hinting at each of the subsections Categorization of approach types. Evolution to recent approaches. Comparison to the proposed solution. | Recent success with gaps in progress. Overall related work approach description, hinting at each of the sections; highlighting gap. Categorization of approach types. Evolution to recent approaches. Comparison of previous approaches to the proposed solution. | Recent success with gaps in progress. Overall related work approach description, hinting at each of the sections; highlighting gap Categorization of approach types. Evolution to recent approaches. Comparison of previous approaches to the proposed solution |
Notes:
- 2 or 3 groups
- Relatively consistent format
- Headers are categories of approaches.
Structure of the Conclusion (+ Broader Impacts)
With the conclusion, we’re going to find similarities with the abstract; in a bit, we’ll also discover a key difference. We’re going to look at the conclusion paragraph-by-paragraph. For each paragraph, once again, let’s write down the question that each sentence answers.
| What is the solution introduced in the paper? What are the key components of the solution (x2)? What are notable results (x2)? What are interesting directions for future work? | What is the solution introduced in the paper? What are the key components of the solution? What are the strengths of the solution? What does the solution point to for the future? --- What does the solution motivate? What are limitations of the work, including identified biases, and how effective are efforts to mitigate them? | What is the solution capable of? What are the possible concerns with the use of the models before they can be deployed x2? --- What is the solution introduced in the paper? What is the strength of the solution? What do the experimental results positively demonstrate? What does the solution motivate? |
- Notes:
- 2 paragraphs.
- Includes a broader impact paragraph.
- Similar to abstract, but includes what the solution motivates.
Structure of the Methods & Experiments
Finally, the most challenging of the sections: the methods and experiments. These have more flexibility in their structure, and grouping them together allows us to see how some of the same elements are reorganized in different ways by papers between the two sections. For each paragraph, we are going to write down the purpose of that paragraph.
| Methods Overall approach description, hinting at each of the sections Describe the architecture and flow of input to output Describe each objective / loss function Experiments Describe the data used Describe the implementation details. Describe usage of model for different tasks For every task type, what results are achieved Describe the ablation experiment and result | Methods Overall approach description, hinting at each of the sections Describe the architecture and flow of input to output Describe each objective / loss function Describe the implementation details. Describe Datasets used. Experiments Overall experiment task setup For every task type, what results are achieved | Methods: Overall approach description, hinting at each of the sections Describe each objective / loss function Describe the architecture and flow of input to output Describe usage of model for different tasks Experiments Overall experiment task setup Describe Datasets used. Describe the implementation details. Describe usage of model for different tasks For every task type, what results are achieved Describe the ablation experiment and result |
- Notes:
- Most elements are the same between papers, but might be found in either methods or experiments.
- Methods include the overall approach description, architecture, flow of input to output, loss function.
- Experiments sections end with describing what results are achieved for different tasks followed by ablations.
Your turn to make a few more notes. Read through the methods and sections to answer the following questions:
- What is the relationship between paragraphs within a section? Hint: think across space and time
- Are any results referenced in the methods, preliminary or not?
- When do results make references to the figures?
- How are previous approaches compared to?
- Can you come up with a checklist for implementation details?
- How many papers are referenced in methods and experiments?
- How are experimental results described?
Resulting Template
不要从空白页开始写论文，而是从相近论文中抽取结构模板
Example template
- Format: Do you know which venue you are formatting the paper for?
- Figures:
- Method overview diagram
- Lower-level methodology diagram
- Comparison of models to other models broken down by task (between 3-5)
- Comparison to ablations
- Abstract
- What is the background for the class of models?
- What is the key gap with previous models?
- What is the key desiderata of a solution?
- What is the solution introduced in the paper and what are notable takeaways?
- What are the components of the solution?
- What are the strengths of the solution?
- What are notable results, mentioning tasks and numbers?
- Introduction
- How successful has the class of approaches under consideration previously been? How do previous approaches in the class tackle the problem?
- What is a solution to the possible approach? What is wrong with that class of approach? What is the key desiderata of a solution?
- What is the solution introduced in the paper? What are the key components of the solution (x 6) ?
- What are the experiments that are performed? What do the experimental results demonstrate? What are the strengths of the solution?
- Methods
- Overall approach description, hinting at each of the sections
- Describe each objective / loss function
- Describe the architecture and flow of input to output
- Describe the implementation details.
- Experiments
- Overall experiment task setup and usage of models for different tasks
- Describe Datasets used.
- Describe the implementation details.
- For every task type, what results are achieved
- Describe the ablation experiment and result
- Related Work
- Recent success with gaps in progress.
- Overall related work approach description, hinting at each of the sections highlighting gap
- Categorization of approach types. Evolution to recent approaches. Comparison of previous approaches to the proposed solution
- Broader Impact and Limitations
- Conclusion
- What is the solution introduced in the paper?
- What are the key components of the solution (x2)?
- What are notable results (x2)?
- What is the strength of the solution?
- What does the solution point to for the future?
- What are limitations of the work, including identified biases, and how effective are efforts to mitigate them?
- What are interesting directions for future work?