Don't have time to read a paper in detail?

Here is how to extract relevant information instantly:
👇 The setup:

You have a rather large and relevant conceptual paper.

There are others and you are not sure if it's worth reading it.

You are however more interested in the "big picture" not the methods and details.

Want to remember every paper you read?

Replace Zotero with a reference map and leverage your spatial memory.

Here is how:
👇

1. Reference manager vs map

Take a look at this screenshot: Which one looks more approachable and interesting? On the right is Zotero displaying all your papers in an endless list. On the left is a reference map.

Reference maps lay out your papers or PDFs in 2D on an endless surface called a "canvas" or "whiteboard". There are many tools that are capable of doing it: Obsidian is an obvious choice, Heptabase is great too, DrawIO is more complex but also good.

2. Zooming in and out / Finding things

Using the scroll wheel or pinch gesture you can fluidly navigate between the bird's eye view and the detailed view with your own notes on a single paper. Left: Detail, Right: All Papers.

To find papers you "fly up" and then "land on" the paper you are looking for. It feels incredibly natural and easier than scrolling through a list.

3. Why it works: Spatial Memory
Humans evolved moving around as hunter gatherers and spatial memory is a key trait needed for navigation. You leverage it by laying out your papers in a landscape, not a list. Your papers gain location and relation.

4. Headers and Topics: Location
Now that your papers are on a landscape or map you can have "countries" on this map. Every country is a topic, further subdivided in sub-topics. Here is the "Machine Learning Country" in the far south west of my map:

I can refer to "papers in the south west" - this is spatial memory being leveraged to remember where things are.

5. Semantic Connections: Relation

The next step is to build the "roads" between locations on the map. Simply draw an arrow and write on it what this relation signifies.

In the above example Swenson 2020 (top) wrote "the trait-demography relationship is weak (Yang 2018)".
So I read Yang 2018 (left) and added a connection. Later I found that (Lynn 2023) suggested a few solutions and linked those two as well.

By just looking at this map you can immediately write a sentence for your literature review. A narrative emerges and synthesis begins.

Summary:

Lay out papers on a spatial canvas using e.g. Obsidian instead of Zotero. Remembering them will be much easier because you can use your spatial memory. Synthesis starts happening automatically when you annotate connections between papers.

Do you do something like this?
Share a screenshot with us!


This is one of the methods you can learn about in my upcoming webinar:

https://twitter.com/Artifexx/status/1767442849438830848

6 Months ago my core tools barely included any AI.

Today things have shifted - a short update to this post:

https://twitter.com/Artifexx/status/1704046981344301339

1. Writing with 3 AI GPTs

When writing I consider 3 things:
- Finding citations to support the argument
- Putting ideas effectively on paper
- Skimming papers for logical relevance

ChatGPT can now do all of these with the help of GPTs:

effortlessacademic.com/a-complete-gui…

Webinar Apr 6th:
Literature Review & Academic Writing with AI

→ Find the most impactful literature quickly
→ Uncover reference gaps
→ Aid your writing process faithfully & ethically with AI

Link:

All details below:
👇effortlessacademic.com/elr3-webinar/ In this video I explain why the major changes in AI are such a big deal for literature reviews:

How do you conduct a lit-review in 2024?

❌ Old way: Searching for papers by keywords on Google
✅ New way: Semantic search using AI tools like SciSpace

Here is a demo:
👇 1. AI Search

Internet search replaced tedious searches in libraries and journals.

Similarly, semantic search will replace the internet search.

AI allows to search answers to questions directly, rather than using keywords.

SciSpace is one such semantic search tool. Let's try!

Overwhelmed with too many things for your paper or project?

Create an academic command center.

Here is how to do it:
👇 1. In @obsdmd install the Kanban plugin

Kanban is a free community plugin.

When I read papers, I often get distracted.

This setup helped to reduce distractions and gain deeper focus:
👇 I use @obsdmd to store notes and PDFs and @zotero to extract and import meta-data. The excellent "Zotero Import plugin" imports papers from Zotero.

Do not use Zotero for highlights because they are stored separately from the PDF.

You can just use the system's PDF viewer.

How important is coding for your scientific work?

As a computational ecologist I code every day, but about 50% of my code is AI.

Here are a few threads that can help you become an AI programmer:
👇 Generally this is my coding workflow:

https://twitter.com/Artifexx/status/1708699305630994927

Don't want to forget what you read?

→ Annotate the PDFs and connect them to your notes.

Here are 4 ways/plugins that make it easy:
👇 The key to remembering is to integrate your annotations into your notes, not just make them.

Read the PDF and highlight, this is ok, but won't help you remember.

Instead, import your highlights into your notes and and connect them together.

This is how it looks:

AI won't create academic writing from scratch.

But used smartly, it's a tool to make your words more coherent, elegant, and fluid.

Follow this example to save a lot of time: 👇

1. I reviewed several papers and extracted key points. But integrating these into a coherent paragraph requires time and skill, particularly for non-native English speakers:

💡 To help with this, I will create an AI assistant. Its task is to combine loose sentences into a paragraph of academic writing.

🤖 Think of AI assistants as a set of additional instructions you add to the normal AI conversation. They make it much easier to automate everyday tasks like this one.

Google or ask AI how academic writing is supposed to look and add your requirements (e.g., I am using markdown to format text, so I want to keep this format).

These rules form the basis of your assistant's prompt. Here is how it looks in my case:

Now I copy and paste the sentences to my newly created AI assistant and get two results to choose from:

(repeat for more alternatives or continue the conversation and ask for changes)

🛑 Do not use this result in your final paper.

It is far from perfect or "academic". AI tends towards generic expressions like "In summary,..." or "Consequentially,..." for example.

⏱️But, it is a massive time-saver as you can now focus on polishing your paragraphs, rather than writing the first draft.

Additionally, you can go back and tweak your assistant step by step. Every time you add something, it gets better and your job is done faster next time.

Btw, the best way to get better results is to add examples to your prompt. This is called "Multi Shot Prompting".

👨‍👩‍👧‍👦I use dozens of AI assistants in my research. It makes my work more enjoyable and more productive.

Summary:

1. Read papers and grab relevant ideas from them.
2. Throw these ideas into a sentence collection
3. Use an AI assistant to join them into a paragraph.
4. Use your words to polish the end result.
5. Feed finished examples back to your assistant to improve results

(Keep in mind that the use of AI sometimes needs to be declared when you submit your paper. )

If you are curious how to create your own AI assistant, follow this tutorial:

https://twitter.com/Artifexx/status/1746442577598112086

Yesterday, I chatted with a friend about writing her MSc thesis.

She said, "It was the worst time of my life because I didn't have a writing process."

Now, I am in her shoes, having to deliver a 10,000-word draft in the next few weeks.

Here is my AI-driven strategy:
👇 In this thread, we will focus on the introduction part of your paper.

AI can significantly speed up the process from here if you have read a few papers and taken some notes.

If you are entirely new to the field, go and read first.

Not confident at coding, but need it for your research?

Here are 8 AI tools that will code for you:
👇 1. @cursor_ai

A standalone editor with a lot of features.

Can be used for free if you do not code much.

Best feature: Auto-Debug → Given an error message will search your code and try to figure out why it crashed.

Overwhelmed by 100s of papers in your lit review?

My newest AI workflow finds the key papers instantly:
👇 1. Export papers

Export papers from Zotero.

Right click on your collection and select "Export Collection..."

Select "BibTeX" as a format.

Save the file to your hard drive.

Mobile GPTs are completely changing how we learn.

Here is how to build an AI assistant to teach you art:
(Who talks to you right there in the museum!)
👇 1. Mobile GPTs = GPTs + Mobile App + Prompt

GPTs allow you to add some instructions and files to ChatGPT to make it accomplish a task with ease.

Using GPTs on the phone is even more powerful:
Snap pictures and have GPT explain/interact with them.

AI scientists are discovering new ChatGPT prompts almost every day.

But only very few people read these papers.

Here are 3 "Prompt Secrets" from recent publications:
👇 1. Recursive Criticism and Improvement (RCI)

Kim 2023 (links in last post) asked ChatGPT to critique its own output.

This critique was then used to create an improved output.

The process is repeated multiple times.

Here are prompt & example on the task of paraphrasing.

Your note-taking app is the most important software in research.

But which one is the best for you?

I compared @Heptabase and @obsdmd. Deep dive in article or key points in thread:

👇
effortlessacademic.com/note-taking-to… @Heptabase is a note-taking tool that strongly focuses on visual notes.

In my opinion: Superb for learning new topics.

Three things it does very well:

- PDF annotations
- Usability/UI
- Whiteboard

Obsidian is the best information management tool for researchers.

One big reason for this are so called aliases – similar apps lack this feature.

Here is how to use it:
👇
An alias is an alternative name for your note.

Why is this a big deal?

Because it allows you to store your notes under various names. Typical use cases:

1. Synonyms
2. Abbreviations
3. Plurals

This makes your notes much more readable and easier to find.

Want to remember every paper you read?

🎥 Here is how you can do it:
(detailed tutorials in the thread )
👇 Upcoming Webinar on Academic Knowledge Management:
- Organize a gigantic knowledge collection
- Synthesize information out of it
- Structure long research projects

https://twitter.com/Artifexx/status/1720737061962186974

Want to finish your research project/PhD 2x faster?

Use this powerful method:
(Key points and templates in thread)
👇 Upcoming Webinar

It is hard to show advanced topics in a few minutes of video here on X.

Better: Join the upcoming webinar. It will make you finish your projects in less time and with better results.

https://twitter.com/Artifexx/status/1720737061962186974

Stuck in your research or lit review?

Ask 100s of fellow researchers for help, data, or feedback.

ResearchHub makes it possible:
(and earn money for helping peers!)
👇 1. Community

Think of ResearchHub as a Quora or StackOverflow for researchers.

But you can not only ask questions but also ask for data, peer review or any other support.

So why would someone help you with any of these for free?

The answer is: ResearchCoin.

Paperpal just released its ChatGPT for academics.

It can generate abstracts, emails to journal editors, and cover letters.

Here is how to use it:
👇 1. New to @teampaperpal?

Given a first draft Paperpal can quickly improve your writing by suggesting numerous edits.

Here is how to use it:

https://twitter.com/Artifexx/status/1661301332446179330