iiinova — where your digital self keeps evolving

Reposted by Zoe LeBlanc

Ted Underwood @tedunderwood.com · 22/09/2026

Models like Talkie-1930 sound like voices from the past. If they could reliably speak from specified historical vantage points, researchers might also use them to simulate the past. But how reliable are they? Today we release a benchmark answering that question for English contexts 1831-1930.

Free-text evaluation of answers to character modeling and constrained generation questions. This is just one of several scores Chronologic-EN can produce; we focus on it here because it's both the hardest test and the one most relevant to simulation of the past. Frontier models reach 72%; Talkie-1930 is stronger than several larger competitors, but not at the frontier by this measure. Note that this score has improved ~45% in the last two years, but still falls perceptibly short of ground truth.

515741

Reposted by Zoe LeBlanc

Amanda Regan, Ph.D. @regan008.bsky.social · 06/08/2026

I have not blogged in probably ten years, but I finished the syllabus for the fourth version of my advanced digital methods course this week and decided to get some thoughts down about teaching digital history in the age of AI. amanda-regan.com/posts/teachi...

3228

Reposted by Zoe LeBlanc

ACH @ach.bsky.social · 15/05/2026

ACH 2026 election results are in! Congratulations to Co-VPs/Presidents Elect Mandy Regan & Brandon Walsh, and new Executive Council members Hannah Alpert-Abrams, Samantha Blickhan, Rachel Starry, and Laura Braunstein. buff.ly/7dg8Ctb

ach.org

Results of the ACH 2026 Elections

2329

Reposted by Zoe LeBlanc

Dr. Jennifer Isasi @jenniferisasi.bsky.social · 13/04/2026

🚨 Join us next week for a conversation with the amazing @zoeleblanc.bsky.social. Have you been wondering how we can pair big DH datasets, translation and AI? She has a great example of that mix 🍲 🗓️ Wednesday, April 22nd at 12PM EDT Register at tinyurl.com/dla-leblanc #DigitalHumanities

Poster for an invited talk at the Digital Liberal Arts Research Initiative at Penn State: “HOW DID THEY DO IT? EXPLORING MULTILINGUAL METADATA AND THE LIMITS OF AI-AUGMENTED TRANSLATION IN DH” by Zoe LeBlanc taking place on April 22nd, 2026 at 12PM EDT.

People can join on zoom.

The poster is mainly blue, with highlights on yellow, and it has a picture of Zoe.

1138

Reposted by Zoe LeBlanc

Ted Underwood @tedunderwood.com · 31/01/2026

This is absolutely how we should pitch coding agents to humanists. Strictly speaking, what mattered about code was never arbitrary syntax, but clearly specifying data structures and processes. And coding agents still force/teach you to do that part. But … shhh! Let’s get ppl on board.

57912

Zoe LeBlanc @zoeleblanc.bsky.social · 26/01/2026

Is anyone else getting some very weird #DH2026 submissions to review? And by weird, I mean almost entirely seemingly AI-generated? Like with entire sections missing text and weird giant fonts and bizarre footnotes. Curious how are other conferences handling this issue??

220

Reposted by Zoe LeBlanc

Melanie Walsh @mellymeldubs.bsky.social · 11/12/2025

Which canonical American authors are the public reading, and why? To find out, we analyzed library borrowing patterns for every author in the Norton Anthology of American Literature (1945 to the Present). Excited to share this new CHR paper & data! anthology.ach.org/volumes/vol0... #CHR2025

116644

Reposted by Zoe LeBlanc

University of Washington Information School @ischool.uw.edu · 04/12/2025

#CFP: Our Center for Advances in Libraries, Museums, and Archives (CALMA) invites proposals for the inaugural Black Information Futures Symposium, February 21–23, 2026, in Seattle. Please save the date & reshare! Submit a proposal by 12/10: #phdchat calma.ischool.uw.edu/uw-news/call...

Lavender graphic with white outline and dark purple text. UW CALMA and the University of Washington Information School logos.

055

Reposted by Zoe LeBlanc

Rebecca Sutton Koeser @suttonkoeser.bsky.social · 01/12/2025

Do you write research software for the humanities? Consider writing a software paper to share your work! Reach out to me, @nolauren.bsky.social , or any of the other editors at the CHR journal with your questions or ideas.

02914

Reposted by Zoe LeBlanc

Lydia Walker @lydiawalker.bsky.social · 10/11/2025

Those interested in a potential roundtable on "Historical Research and Artificial Intelligence: Critical Reflections on Method, Power, and Possibility in International History" that @zoeleblanc.bsky.social and I are planning for @shafrhistorians.bsky.social 2026 - please send us your abstracts!

134

Reposted by Zoe LeBlanc

Ryan Cordell @ryancordell.org · 06/11/2025

Job alert! @ischoolui.bsky.social is hiring in 4 areas this year Early literacies: illinois.csod.com/ux/ats/caree... Information behavior/HCI/UX: illinois.csod.com/ux/ats/caree... Information, Culture & Society: illinois.csod.com/ux/ats/caree... Open IS: illinois.csod.com/ux/ats/caree...

13029

Reposted by Zoe LeBlanc

Yuerong Hu @yrhu.bsky.social · 18/08/2025

My co-authors, Jana Diesner, @tedunderwood.me, @zoeleblanc.bsky.social, @gworthey.bsky.social and @profdownie.bsky.social, and I are excited to share our paper in @bigdatasoc.bsky.social "Who decides what is read on Goodreads?" on book review sponsorship, open access at doi.org/10.1177/2053....

Hu, Y., Diesner, J., Underwood, T., LeBlanc, Z., Layne-Worthey, G., & Downie, J. S. (2025). Who decides what is read on Goodreads? Uncovering sponsorship and its implications for scholarly Research. Big Data & Society, 12(3). https://doi.org/10.1177/20539517251359229 (Original work published 2025)

35720

Reposted by Zoe LeBlanc

Ryan Cordell @ryancordell.org · 13/06/2025

#ThanksBrett for always looking forward. You saw a bright future for the humanities, as key interlocutors in the most pressing socio-technical discussions of our moment, and you created opportunities for those conversations to actually happen. We all owe @brettbobley.bsky.social enormous debts

1385

Reposted by Zoe LeBlanc

Meredith Martin @mmvty.bsky.social · 22/04/2025

Today's the day! I invite close readers to see if they can find the places where I channeled the most rage into this project. It is supposed to make you uncomfortable. You are supposed to try to dismiss it as "praxis" vs. "theory" and then you are supposed to be convinced otherwise.

811136

Zoe LeBlanc @zoeleblanc.bsky.social · 16/04/2025

Dumb Q but genuinely need help: say you have a few weeks (maybe only 2) to build a website what's your dream tech stack? Ideally deploy from GitHub though could white knuckle it through AWS, custom domain, & will have complex data viz (d3 & vega) and navigation;

361

Reposted by Zoe LeBlanc

Melanie Walsh @mellymeldubs.bsky.social · 11/04/2025

I missed this announcement back in March. "For Schmidt Sciences, Digital Delacroix is the first of a projected 10 to 15 grant recipients that will receive a total of $10 million to apply A.I. to research in the humanities." www.nytimes.com/2025/03/26/a...

nytimes.com

New A.I. Project Explores Mysteries of Delacroix, Master of Romanticism (Gift Article)

Eric and Wendy Schmidt and the Sorbonne will fund a new program to digitize Delacroix’s papers and identify other artists who may have contributed to his murals and paintings.

081

Reposted by Zoe LeBlanc

Alexander Doria @dorialexander.bsky.social · 11/04/2025

Pre-announcing one of the first fork of Common Corpus: French ScienceCommons, the largest open corpus of French academic publication (about 30 billion words). New Pleias project unveiled on Wednesday at the French Académie des Sciences.

1145

Reposted by Zoe LeBlanc

Jonathan Cheng @jonathancheng.bsky.social · 11/04/2025

So strange, as the stress and worry about the world grows in this household, so to do the number of Japanese wood-block prints. In order: Hiroshi Yoshida, Shiro Kasamatsu, and Hasui Kawase

5264