26 June 2025

Link roundup for June 2025

Let’s see if I can get back into a regular rhythm of posting these!

• • • • •

There continues to be a small group working on creating posters by computing methods.

An human generated poster on the left and a computer generated poster on the right of the same content.

A new preprint by Pang and colleagues not only dives into poster creation, with a program called PosterAgent, but how to evaluate computationally generated posters, called Paper2Poster. Existing poster assessment rubrics are not good for evaluating computationally generated posters, because the computer makes mistakes that humans don’t make.

(O)ur Paper2Poster pipeline, built on a fully open-source toolbox... surpasses existing GPT-4o–based multi-agent approaches on nearly all metrics while consuming 87% fewer tokens.

While they claim their Paper2Poster assessment rates posters similarly to humans, it’s worth noting their comparison is based on one human rating five posters.

Their technique involves converting completed academic papers to posters. Since most posters are made well before a paper is ready for publication, we are still a long way from something that will be useful to most poster makers. And I’m not sure large scale automation for poster creation is necessary.

Pang W, Lin KQ, Jian X, He X, Torr P. 2025. Paper2Poster: Towards multimodal poster automation from scientific papers. arXiv: arXiv:2505.21497. https://arxiv.org/abs/2505.21497

• • • • •

Benedikt Ehinger  has created Illustrator templates for conference posters that are free to use. Here’s an example:

A conference poster about "filled in precepts" in landscape orientation.

Because it’s in Illustrator, there are a couple of features that you might not find in other static templates. 

  1. Box backgrounds adjust for titles.
  2. Borders can be easily resized; the curves don’t change with rescaling.

• • • • •

Staying with Benedikt for a moment, he also makes art from his graduate students’ theses. Examples:

Poster showing three wave-like patterns created from letters.


 The plot above is made from the first 10.000 letters of the student’s thesis.

Pages of text with a few areas spotlighted.

Benedikt writes:

The idea of “thesis art” is to inspire discussion with persons who do not have an academic background or work in a different field. The thesis is hidden in the drawer, but the poster is out there at the wall for everyone to see. 

• • • • •

If you are teaching using conference posters, Stefanile has a paper that describes how to teach people how to present posters:

(T)he approach of preparing a poster, emphasized here, involved a collaborative effort between the presenter and their colleagues/mentor to properly prepare the poster, particularly emphasizing the use of consistent best practices when presenting the poster. 

Stefanile A. 2025. Designing and presenting an oral poster presentation for undergraduate students’ research. Natural Sciences Education 54: e70018. https://doi.org/10.1002/nse2.70018 

• • • • •

Sticking with teaching, Nicola Koyama has built in some self-assessment practices that help students engage with the material. Excerpt:

Some of Nicola’s students aren’t keen at first, she says “I think they grumble having to do something. But then at the end... they’re excited to know whether they've judged their poster correctly.”

Koyama N. 2025. Know where you atand: How incentivized self-assessment transforms poster presentations. Liverpool John Moores University Student Experience Proceedings Special edition: Case studies in assessment and feedback design: https://doi.org/10.24377/studentexp3175

• • • • •

I have frequently reference Nate Piekos and Blambot, one of my favourite supplier of comic style fonts, but I am not sure I have referred to his page of compiled lettering tips.

The 94% line cheat - taking a very long line and compressing it slightly so it fits better,

This line cheat may be helpful in some poster layout situations, where you might be able to gain a line back to make text fit into allotted spaces better.

• • • • • 

This next paper doesn’t mention posters directly, but is one of the best discussions I have ever read about the challenges of holding an online conference that feels like a conference and not a Zoom call. 

The problems are partly organizational. The organizers wanted real time events, not asynchronous ones. This meant figuring out events for many time zones.

The problems are partly technical. What software exists that can make people feel like they re in physical spaces that encourage the equivalent of hallway conversations? They describe an app called QiQoChat, which I have not heard of before.

The organizers also realized that they needed events that would bring people together. For one, they live music. 

Jonathan Foster (right) and David Lennon (left). Members of the Edinburgh University Folk and Traditional Music Society at the conference ceilidh.

This paper encourages me to think that a good online poster session is possible. Recommended.

Bastian M, Flatø EH, Baraitser L, Jordheim H, Salisbury L, van Dooren T. 2022. ‘What about the coffee break?’ Designing virtual conference spaces for conviviality. Geo: Geography and Environment 9(2): e00114. https://rgs-ibg.onlinelibrary.wiley.com/doi/abs/10.1002/geo2.114

• • • • •

In two gynecology conferences, 46.8% of oral presentations were published compared to just 14.2% of poster presentations. As is common, there is no discussion of why this difference exists or whether tany steps should be taken to try to increase the amount of publications deriving from posters.

Rotem R, O’Leary BD, McCarthy CM, O’Reilly BA, O'Sullivan OE. From conference to journal: Analyzing the path to publication for EUGA and JOGS 2022 abstracts. International Journal of Gynecology & Obstetrics: in press. https://doi.org/10.1002/ijgo.70319

• • • • •

This conference report makes me regret that I wasn’t able to attend Woodstock of Biology2 + Night Science (#TCTeAC on Bluesky).  Presenters got walk-up music, there was a wedding (!), questions people were too embarrassed to ask, and much more. Seen from a distance, it sounds like it feels like Science Online meetings used to. It sounds like a lot of fun was had on top of the science.

But. 

Poster session in the forest — because… why not?

A bunch of poster boards among large gree trees.

 (Photo from Barak Rotblat on Bluesky.)

Although it is apprently a volleyball court in a forest.

Outdoor poster session.

(Photo by Nir Eynon.)

Poster session with trees visible in back.

(Photo by Zoe Harrington on Bluesky.)

Poster at Night Science meeting

(Picture from Claire on Bluesky.)

Um. I, personally, can think of several reasons why not. I know some people would see one tiny woodlands creature and be, “THERE’S A BUG THAT’S IT I’M OUT.” 

Or slugs, for that matter. This photo...

Woman with small string instrument standing next to poster outside.

Was accompanied by the text, “INNAPROPRIATE SLUG SONGS.” (Photo by ‪Valeriy Kutsyna.) 

And posters were burned

Posters being burned at NIght Science conference.

(Photo again from Claire on Bluesky.)

Not sure I can get behind that. Even if if it was voluntary.

The conference ended when police showed up.

• • • • •

And that’s all for now! Thank you, as always, for your attention! 

19 June 2025

Critique: Computing crocetin

This week's is courtesy of the extremely patient Dimitra Bezantakou. Dimitra was nice enough to say that she found the blog helpful as she was creating the poster below. Click to enlarge!

Poster titled, "In silico evaluation of Crocetin as a potential inhibitor of Α β oligomerization in Alzheimer’s disease."

This is a common three column poster, but there are a few things that help set it apart.

Bolding key words and phrases in the introduction makes reading the text faster. The text has ample spacing, which also helps readability.

The methods are shown in a way that add some visual interest to the poster. There is a lot of text and anonymous graphs, so the blocks of colour were quite badly needed to stop the poster from being too monotone.

The title is the biggest thing on the poster. And while I am not excited about the logo to the left of the title, the subdued colour and size stop it from distracting from the title.

Wider margins might improve the look of the poster. As clean and orderly as the columns are, the space between the columns looks extremely tight. The graphs in the centre column are just too close to the text on either side. 

Here is a quick revision with the central graphics reduced to 95% of their original size:

Poster titled, "In silico evaluation of Crocetin as a potential inhibitor of Α β oligomerization in Alzheimer’s disease," with slightly smaller graphs in the central column.

This helps. Probably the same approach could be used to bring the text away from the edges of the page, which are also closer than readers might expect.

13 June 2025

How to pitch a poster in lightning talk (ePosters blog guest post)

The good people at ePosters asked if I would write something for their blog, They asked nicely, so I said, “I’ve been meaning to write about flash talks where people pitch their posters. How about that?” They said, “Okay.” 👍 And now it’s done and up.

Please visit ePosters for “How to pitch your poster in a flashy lightning talk.”

I hope you enjoy!

12 June 2025

Poster controversies

Over a year ago, I saved a few tweets about controversies that erupted over a couple of posters at a Society for Personality and Social Psychology conference. At the time, I was aware that there was a lot of discussion on social media but didn’t have the bandwidth to follow it closely.

Today I went looking for summaries of the controversy, and found a couple, which I have linked below. I don’t pretend that these summaries are complete or perfect, but I think they give at least a hint of what happened.

Most conferences do minimal screening of conference posters. They also do not necessarily do great monitoring of what goes up on unused poster boards, or whether people post the work they originally submitted. Normally, these are harmless.

But I do think it is worth asking under what conditions posters should be removed by conference organizers. If an organization doesn’t have policies for that, they should make them, and make them public, rather than trying to react on the fly.

External links

The censorious Society for Personality and Social Psychology (13 February 2024)

SPSP censorship 2.0: “We are above the rules” (21 February 2024)

SPSP addresses convention poster concerns and outlines next steps (29 February 2024)

 

05 June 2025

Critique: Posters have entered the chat

Some time ago, Greg Fell shared an interesting poster. (You may not be able to see it due to user settings.) I reached out to the creator, Anton Pottegård. He sent a slightly earlier version of this poster. Click to enlarge!

Poster formatted as a chat in a phone messaging system.

This poster does a couple of things. First, it draws inspiration from an existing design that people already understand. Second, it is memorable because it breaks from the conventional “journal article on a page” format. Others have used the text message style to create concert posters.

I don’t know exactly how this particular poster was created, but it is not hard to find websites that can emulate a chat.

What is not clear to me from this version is how someone is supposed to get the abstract mentioned at the bottom of the poster. I appreciate that the PDF icon looks like what you would see in a text message, but a short, easy to remember URL might have done the job, too.

Is this format appropriate to all posters? I don’t think so. This study has a simple result: “There’s no effect.” Many other projects are nowhere near as easy to summarize.

I am sure that some people will argue that this format is never appropriate for an academic poster, for two reasons. 

First, it refers to data but does not show those data. While I understand the reasoning behind the “I must be able to personally inspect every data point before I can be convinced,” I don’t think that level of intense scrutiny is at all necessary for a conference poster. Save it for journal articles. 

Second, there are no references. I am neutral on the subject, but forthcoming research shows that people think less of posters without at least one reference.

Bonus posters 

Anton wrote, “Overall, doing strong posters is very important to me!” He shared more that he created.

Here we have one that is very different in style from the one above:

Handdrawn poster on antipsychotics and breast cancer

While the first poster imitates the glossy interface that almost everyone has become familiar with, this second poster uses a hand drawn style. I like the style, but I worry that the summary in the top half is a little hard to read.

(The QR code at the bottom still works!)

And this is one of Anton’s favourites:

Poster of person on toilet reading a paper that says, "Use of proton pump inhibitors continues to rise."

Same style as above, but the summary is shorter, larger, and more readable.

While this poster may have been on on Anton’s favoruites, I suspect it would not be everyone’s cup of tea. I personally wouldn’t make a poster that included someone on a toilet, but different strokes for different folks.

Pros: Memorable.

Cons: I don’t see that it as relevant to the content. If this was a paper about gastrointestinal function, I wouldn’t question it is much.

Related posts

Critique: Protein biosynthesis 

External links

Anton Pottegård home page

Sample posters by Anton