# Explaining Explainpaper

**URL:** https://connect.oeglobal.org/t/explaining-explainpaper/4416
**Category:** The Plaza
**Tags:** ai
**Created:** [November 14, 2022, 9:13pm UTC](https://connect.oeglobal.org/t/explaining-explainpaper/4416 "2022-11-14T21:13:43Z")
**Posts on this page:** 1
**Showing post:** 1

<div class="post-metadata">

### Author: ![cogdog](https://connect.oeglobal.org/user_avatar/connect.oeglobal.org/cogdog/32/2_2.png) [@cogdog](https://connect.oeglobal.org/u/cogdog)
#### Post date: [November 14, 2022, 9:13pm UTC](https://connect.oeglobal.org/t/explaining-explainpaper/4416/1 "2022-11-14T21:13:44Z")

</div>

In my [ongoing dabbling to understand what some call AI](https://connect.oeglobal.org/t/understanding-doing-some-ai/4013) (@poritzj sez no) I explored [a bit of Elicit](https://connect.oeglobal.org/t/eliciting-effectiveness-of-elicit-ai-research-lit-review-tool/4191), a tool aimed at helping locate research papers with AI

In a similar vein comes [Explain Paper](https://www.explainpaper.com/), which has almost no information about itself besides “Upload a paper, highlight confusing text, get an explanation. A better way to read academic papers.”

But rather than tweeting and running, I like to try things out, first with their sample papers. From what I could see the typical use is highlighting a sentence in a paper, and let the site rephrase it in a more understandable manner. Which it did, following that, you can enter a followup question.

To give it a test, I uploaded a PDF of a recent IRODDL paper [Impact of COVID-19 on Formal Education: An International Review of Practices and Potentials of Open Education at a Distance](https://www.irrodl.org/index.php/irrodl/article/view/6120) with many co-authors who are here in OEG Connect and presented this work at our conferences before publication.

Keep in mind, I just tried a few things to see what happened. I selected a sentence that describes the research method

> First, the approach uses researcher triangulation (Denzin, 1978) to frame research  
> questions which further increases the reliability and validity of the research (Creswell, 2012; Oppermann, 2000)

and got a slightly rephrased explanation, and it was within seconds.

Then I asked a follow up question that was not explicit in the paper, “How do we know the triangulation method is valid?”

 ![Example of highlighting a passage and asking a followup](https://connect.oeglobal.org/uploads/default/original/2X/e/effd81f77748eb046e2cf6c76e43961f2ef884ee.jpeg)

For fun, I tossed in this question: “Are MOOCs the best way to learn online?”

 ![Response to question -- Are MOOCs the best way to learn online?](https://connect.oeglobal.org/uploads/default/original/2X/7/76c435124d7ce4f3479a57bfc7f5b0c892b23eb3.jpeg)

I noticed after a while, any question asked about “the best way” starts with “There is no one-size-fits-all answer to this question…”

I could not resist putting in the irrelevant question, “What is the best condiment to put on a hamburger?”

 ![Response to question -- What is the best condiment to put on a hamburger?](https://connect.oeglobal.org/uploads/default/original/2X/d/d69eb7a1e41284c75b3edf2efb4102ec126d903a.jpeg)

Again, a question to keep in mind- is this intelligence or is this a statistical response model?

Here is another summary for a sentence describing the UNESCO Recommendation on OER:

 ![Explanation of UNESCO OER Recommendation](https://connect.oeglobal.org/uploads/default/original/2X/6/6b9f2c89c430fd692fa969f33c748f2fc8aca59c.jpeg)

One might get their hankles up because this is rephrasing the papers sentence slightly different. Do we run for the proctoring software here?

I tried one more sentence in the paper about the summary of OER use on Turkey, along with a followup.

 ![turkey-oer](https://connect.oeglobal.org/uploads/default/original/2X/5/559c9f12e7206a959892f5440f0e1d6dc46a8c5d.jpeg)

I find the way this is working interesting, and maybe I should have tried a more complex paper? One might say this paper is so well written no AI is needed!

Now I think I should try a paper that explains AI as I have not found anything tha explains it un understandable fashion that does not just get vague and black box mystery like.

Maybe someone else can give Explain Paper a try and see what kind of results you get.

---

_[View the full topic](https://connect.oeglobal.org/t/explaining-explainpaper/4416)._
