Boris Hejblum
LightNews LightNews
banner
borishejblum.science
Boris Hejblum
@borishejblum.science
41 followers 190 following 0 posts
INSERM researcher in biostatistics & #rstats enthusiast - views are my own
https://borishejblum.science
Posts Replies Media Videos
Reposted by Boris Hejblum
posit.co
Introducing bluffbench, a new tool to evaluate how well LLMs actually see data plots.

When we trick LLMs with secret #RStats transformations, they can miss the visual contradiction.

bluffbench helps us measure this "blind spot" in AI coding agents. Learn more: posit.co/blog/introdu...
When plotting, LLMs see what they expect to see - Posit
Data science agents need to accurately read plots even when the content contradicts their expectations. Our testing shows today's LLMs still struggle here.
posit.co
November 19, 2025 at 4:55 PM
Reposted by Boris Hejblum
jennybryan.bsky.social
Will definitely include this example in my next talk on how to name files!

✅ Full marks for "make it easy to guess what the heck something is, based on it name".
earlymodernjohn.bsky.social John Gallagher @earlymodernjohn.bsky.social · Oct 30
Genuinely delighted to download a PhD thesis from a university repository where the author has neglected to remove the words "BITCH THIS IS YOUR THESIS" from the filename.
October 31, 2025 at 8:46 PM
Reposted by Boris Hejblum
ivelasq3.bsky.social
I loved discussing "Positron for RStudio Users: A Gentle Introduction" with @simisani.bsky.social & R-Ladies Gaborone!

Check out the recording and materials:

📹 www.youtube.com/watch?v=2fOQ...
📝 ivelasq.rbind.io/talk/positro...

I hope this intro is better than the one between my three cats 😹
Positron for RStudio Users: A Gentle Introduction with an image of a cute orange kitten sniffing a hand
October 27, 2025 at 10:20 PM
Reposted by Boris Hejblum
lcolladotor.bsky.social
"What Makes a Good Data Visualization?" 🤔

Check @christinezhang.bsky.social's 🤩 excellent 💯 guest lecture for 140.776 Statistical Computing at @johnshopkinssph.bsky.social

📽️: youtu.be/SeLucCb05Dk

Slides: speakerdeck.com/lcolladotor/...

#RStats @jhubiostat.bsky.social @lieberinstitute.bsky.social
[2025-10-16] What Makes a Good Data Visualization?
YouTube video by Leonardo Collado Torres
youtu.be
October 24, 2025 at 7:22 PM
Reposted by Boris Hejblum
weare.rladies.org
If you haven't got a website and want to give #quarto a go, the first video in my "Quarto Websites" series with @emilhvitfeldt.bsky.social could be a good place to start:
www.youtube.com/watch?v=l7r2...
More resources in 🧵
Quarto Websites 1: Build your homepage | Charlotte Wickham & Emil Hvitfeldt | Posit
YouTube video by Posit PBC
www.youtube.com
October 20, 2025 at 9:51 PM
Reposted by Boris Hejblum
weare.rladies.org
The best I've managed is to use citations as usual, then edit a CSL file to make the style you would normally see in the "References" section, the style you get inline.

Code example: github.com/cwickham/ref...
Preview: cwickham.github.io/references/
October 23, 2025 at 3:28 AM
Reposted by Boris Hejblum
clauswilke.com
We need to have a conversation about random seeds. Don't use 42.
blog.genesmindsmachines.com/p/if-your-ra...
If your random seed is 42 I will come to your office and set your computer on fire🔥
Figuratively. More likely you'll get a stern talking to.
blog.genesmindsmachines.com
October 22, 2025 at 12:49 PM
Reposted by Boris Hejblum
evaheinz7.bsky.social
The linked course is an incredibly well-written, clear explanation of how LLMs work and outlines really thoughtfully what the can, and can't, do. Recommended read for absolutely everyone out there. #MicroSky #MicrobiomeSky 💻🧬
carlbergstrom.com Carl T. Bergstrom @carlbergstrom.com · Oct 18
So why do I let AI in my house? Don’t I care about the environmental consequences? What about the theft of intellectual property? Don’t I understand that these are bullshit machines?

Yes, yes, yes, and yes.

It’s what I do all day long.
Modern-Day Oracles or Bullshit Machines: Introduction
A free online humanities course about how to learn and work and thrive in an AI world.
thebullshitmachines.com
October 19, 2025 at 9:01 AM
Reposted by Boris Hejblum
andrew.heiss.phd
can't wait to do my vibe research and vibe writing and vibe analysis and eventually get vibe promoted
October 1, 2025 at 5:00 PM
Reposted by Boris Hejblum
thomasp85.com
🎨 Theming got a huge overhaul with the latest #ggplot2 release. In honour of that @teunbrand.bsky.social has written a comprehensive deep-dive into styling your plots, covering both old and new functionality. Grab a coffee and dive in!

#rstats
ggplot2 styling
This post discusses one function in ggplot2: `theme()`. Find out about the glamour of graphics in this deep-dive article.
www.tidyverse.org
October 1, 2025 at 8:10 AM
Reposted by Boris Hejblum
vincentab.bsky.social
{tinytable} 0.14.0 for #RStats makes it super easy to draw tables in html, tex, docx, typ, md & png.

There are only a few functions to learn, but don't be fooled! Small 📦s can still be powerful.

Check out the new gallery page for fun case studies.

vincentarelbundock.github.io/tinytable/vi...
a table about lemurs a table about students and schools a table about wines
September 29, 2025 at 12:44 PM
Reposted by Boris Hejblum
collinberke.bsky.social
"The real reason for all this cleaning work is the signal-to-noise ratio in raw data is too poor for purpose we intend. We need to improve data quality to amp the signal for our tools to find."

3/9
September 28, 2025 at 4:59 AM
Reposted by Boris Hejblum
emilhvitfeldt.bsky.social
I'm exited to announce a new resource about making slides with quarto and revealjs. This book is the combination of all the work I have done in this area, reordered and polished up

There isn't a lot of new information yet, but this format allows me to add more easily

slidecrafting-book.com
#quarto
Screenshot of first page of slidecrafting-book.com website
September 24, 2025 at 4:12 PM
Reposted by Boris Hejblum
thomasp85.com
I am super hyped to finally share the first release of plumber2 with all of you. This has been the center of my attention for a big part of 2025 and I hope you'll find it a worthy update to the venerable plumber package.

The blog post will tell you more

#rstats
plumber2 0.1.0
plumber2, a complete rewrite of plumber, has landed on CRAN, providing a modern, future proof solution for creating web servers in R. Read all about the new features here.
www.tidyverse.org
September 24, 2025 at 6:52 AM
Reposted by Boris Hejblum
emilyriederer.bsky.social
Really insightful post from Julie Tibshirani (spotted in LinkedIn, can't find on Bsky) reflecting on #rstats 's unique governance structure and what can be learned for other languages

jtibs.substack.com/p/if-all-the...
If all the world were a monorepo
The R ecosystem and the case for extreme empathy in software maintenance
jtibs.substack.com
September 14, 2025 at 11:29 PM
Reposted by Boris Hejblum
thomasp85.com
I am beyond excited to announce that ggplot2 4.0.0 has just landed on CRAN.

It's not every day we have a new major #ggplot2 release but it is a fitting 18 year birthday present for the package.

Get an overview of the release in this blog post and be on the lookout for more in-depth posts #rstats
ggplot2 4.0.0
A new major version of ggplot2 has been released on CRAN. Find out what is new here.
www.tidyverse.org
September 11, 2025 at 11:20 AM
Reposted by Boris Hejblum
libbyheeren.bsky.social
People keep telling me they can't afford to attend #PositConf2025 next week, so I need to make a public service announcement: attending virtually is very affordable (or free!) & is amazing! Get registered & hangout with me on the Discord server!! #databs #rstats #python
Screenshot from https://posit.co/conf-pricing-2025/ showing the posit conf virtual ticket pricing. $99 for standard virtual ticket, $49 for Government and Non-profit employees, and Free for students, academics, and needs-based attendees. There are hand-drawn doodles by Libby that highlight each option and one note says "Plus, Virtual Day!"
September 9, 2025 at 7:32 PM
Reposted by Boris Hejblum
arthurhughes27.bsky.social
Glad to share our new open-access publication in Statistics in Medicine of a non-parametric method for the identification of high-dimensional surrogate markers written with Layla Parast, Rodolphe Thiébaut, and @borishejblum.science (doi.org/10.1002/sim....).
RISE: Two‐Stage Rank‐Based Identification of High‐Dimensional Surrogate Markers Applied to Vaccinology
In vaccine trials with long-term participant follow-up, it is of great importance to identify surrogate markers that accurately infer long-term immune responses. These markers offer practical advanta....
doi.org
September 8, 2025 at 9:34 AM
Reposted by Boris Hejblum
posit.co
Introducing Databot: an AI assistant for exploratory data analysis in #Python and #RStats!

A research preview in Positron, Databot is a tireless pair programmer to help you explore data.

Learn more about this tool and our philosophy behind it:

🤖 posit.co/blog/introdu...
⚠️ posit.co/blog/databot...
Collaborate > automate. Databot logo alongside screenshot of Positron.
August 29, 2025 at 2:03 PM
Reposted by Boris Hejblum
andrew.heiss.phd
New #QuartoPub extension! Quarto output styling adds CSS rules for computational output, warnings, and messages for #rstats, #python, #julialang, and #observablejs - use the default styles, minimal styles, or add your own custom CSS andrewheiss.github.io/quarto-outpu...
Output, errors, warnings, and messages in R, but they're all formatted the same Output, errors, warnings, and messages in R with a thin border around the output, and errors filled with red, warnings filled with yellow, and messages filled with blue Warnings and messages highlighted with actual R code showing a ggplot plot of penguin data Output, errors, warnings, and messages in R with a thin border around the output, and errors bordered with red, warnings bordered with yellow, and messages bordered with blue
August 25, 2025 at 6:02 PM
Reposted by Boris Hejblum
charlespiller.bsky.social
Impressive look at the impact of paper mills on one journal, by Rene Aquarius, @elisabethbik.bsky.social, David Bimler, Morten Oksvold, @thatsregrettab1.bsky.social h/t @retractionwatch.com www.tandfonline.com/doi/full/10....
Tackling paper mills requires us to prevent future contamination and clean up the past – the case of the journal Bioengineered
Taylor & Francis journal Bioengineered has been targeted by paper mills. Our goal is to identify problematic articles published in Bioengineered during the period 2010 to 2024. Dimensions was used ...
www.tandfonline.com
August 25, 2025 at 3:03 PM
Reposted by Boris Hejblum
dingdingpeng.the100.ci
Ever stared at a table of regression coefficients & wondered what you're doing with your life?

Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social)
Preprint: doi.org/10.31234/osf...
Website: j-rohrer.github.io/marginal-psy...
Models as Prediction Machines: How to Convert Confusing Coefficients into Clear Quantities Abstract Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as “counterfactual prediction machines,” which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models). Figure illustrating model predictions. On the X-axis the predictor, annual gross income in Euro. On the Y-axis the outcome, predicted life satisfaction. A solid line marks the curve of predictions on which individual data points are marked as model-implied outcomes at incomes of interest. Comparing two such predictions gives us a comparison. We can also fit a tangent to the line of predictions, which illustrates the slope at any given point of the curve. A figure illustrating various ways to include age as a predictor in a model. On the x-axis age (predictor), on the y-axis the outcome (model-implied importance of friends, including confidence intervals). Illustrated are 1. age as a categorical predictor, resultings in the predictions bouncing around a lot with wide confidence intervals 2. age as a linear predictor, which forces a straight line through the data points that has a very tight confidence band and 3. age splines, which lies somewhere in between as it smoothly follows the data but has more uncertainty than the straight line.
August 25, 2025 at 11:49 AM
Reposted by Boris Hejblum
emilhvitfeldt.bsky.social
Happy to announce ✨quarto-revealjs-editable✨

This fully supersedes the imagemover extension, as I back then didn't realize the potential. You can now also move, resize, change font size and alignment for text in your slides

github.com/EmilHvitfeld...
#quarto #slidecrafting
August 20, 2025 at 5:39 PM
Reposted by Boris Hejblum
njtierney.bsky.social
Make errors more accessible and better with #rstats package {more} by @milesmcbain.bsky.social - inspired by rustc's detailed error explanations! This idea feels so good.

github.com/milesmcbain/...
GitHub - MilesMcBain/more: Serve up more when things go bad
Serve up more when things go bad. Contribute to MilesMcBain/more development by creating an account on GitHub.
github.com
August 18, 2025 at 4:28 AM
Reposted by Boris Hejblum
itaiyanai.bsky.social
The difference between doing a project and presenting it. An observation can lead to many avenues of explorations before focus turns to a specific discovery. Presenting it, in a talk / paper, follows inversely, with broad perspectives coming before & after the specific discovery.
August 18, 2025 at 3:43 AM