Love this story about how Wikipedia fends off censorship and disinformation: "Wikipedia’s first and best line of defense is to explain how Wikipedia works." www.theverge.com/cs/features/...

Created a small tool for updating a Slack channel when new papers are added to a Zotero library. Totally free using Google Cloud Run's basic tier. levon003.github.io/2025/09/19/z...

A quick love letter to the Internet Archive's free online book loans: it makes it so easy to verify extremely obscure facts! I'm constantly using it to verify info on Wikipedia.

Thinking again about AI as tarmac. (Here's Fred Turner, writing in 2017.) levon003.github.io/2024/05/19/m...

This genuinely shocked me. A Qualtrics heatmap included in a 2019 Nature Comms paper, turned into a far-right meme by the Department of Homeland Security. levon003.github.io/2024/07/27/m...

Starting to get excited for #CSCW2025! I have a paper on peer support via recommender systems in the virtual session. Recommender systems should treat people matching as a serious design goal! levon003.github.io/2025/02/25/c...

The @acm-cscw.bsky.social Starter Pack is growing ahead of #CSCW2025 ! (go.bsky.app/SPumuMT) 🔥 A warm welcome to all of our new members 🔥: @imanm02.bsky.social @jesanovi.bsky.social @shadinz.bsky.social @jaytee-ess.bsky.social @jaewonk.bsky.social @sypark9.bsky.social New member thread 1/x

This article was a supreme delight: grab a cup of tea, set aside 10 or 15 minutes, and enjoy this story about The Hardest Working Font in Manhattan. aresluna.org/the-hardest-... (via @ferociousj.bsky.social)

Very interesting bill. I wrote up a few initial thoughts, including on the peer support exemptions. levon003.github.io/2025/08/11/i...

Amazing things are happening in the UK www.ft.com/content/6256...

I wrote about how AI chatbots strip cultural and historical context away from language. What remains is merely a ghost in the machine, and it's repeatedly making people confused and scared www.wired.com/story/chatgp...

This post tries to explain why I find language models exciting. But it doesn't try to persuade skeptics that they should agree. It's aimed more at people already working with AI, and its goal is to sharpen our collective sense of what the upside potential might be. #MLSky 🤖 🧪

Interesting method. The author (Will Schulz) has a working paper for this ICA 2025 presentation: willschulz.com/wp-content/u...

Validity as a style guide for science.

Short story recommendation: "How to Raise a Kraken in Your Bathtub" by @pdjeliclark.bsky.social. Reminded me a lot of Fallen London from @failbettergames.com; the same number of tentacles, but more colonialism and less hope. www.uncannymagazine.com/article/how-...

Good excuse to re-share "But is it rigorous? Trustworthiness and authenticity in naturalistic evaluation". Written by Yvonna Lincoln and Egon Guba in 1986, so surely we've sorted out how to evaluate varied research methods in the last 40 years. hlanthorn.com/wp-content/u...

Should LLMs be used to review papers? AAAI is piloting LLM-generated reviews this year. I wrote a blog post arguing that using LLMs as reviewers can have bad downstream consequences for science by centralizing judgments about what constitutes good research. bryanwilder.github.io/files/llmrev...

People keep plugging AI "Co-Scientists," so what happens when you ask them to do an important task like finding errors in papers? We built SPOT, a dataset of STEM manuscripts across 10 fields annotated with real errors to find out. (tl;dr not even close to usable) #NLProc arxiv.org/abs/2505.11855

Very swaggy. "By day, you can see scanners scanning and what they're scanning, and when they're not scanning, you will see public domain silent movies and pictures. Sorta like an Internet Archive TV" www.youtube.com/watch?v=aPg2...

Wrote about a fun research controversy involving the finding that Irish judges are more likely to cite legal cases in rulings if they have a Wikipedia article. levon003.github.io/2025/05/21/w...

@xuanalogue.bsky.social Congrats on your recent defense! Totally unrelated question: are you familiar with "Three Nightmares of the Inductive Mind" by Reza Negarestani? It reminded me vaguely of "Beyond Preferences", and I was wondering if it was taken seriously. www.glass-bead.org/research-pla...

"These Are A Few of My Favorite Things" That DAGs don't do: feedbacks, bias, private information, perverse incentives, moderation, and telling you where to cut the graph. wildetruth.substack.com/p/these-are-...

This is really cool (h/t @arnicas.bsky.social's newsletter) madebyoll.in/posts/world_...

I thought this was surely an exaggeration, but it's not! Wow! "Between 1993 and 2006, estimates say that AOL sent out more than 1 billion CDs with free trial software." www.vox.com/2015/5/12/85...

My team got a nice write-up in a blog post from @thelearningagency.bsky.social. I can't complain about effusive comments from my boss! the-learning-agency.com/the-cutting-...

I wrote something up for AI people who want to get into bluesky and either couldn't assemble an exciting feed or gave up doomscrolling when their Following feed switched to talking politics 24/7.

Our lecture topic in our ML eval class today. ✨ Evals have always played an important socio technical role in policy decision-making and surprisingly little changes when you add "AI" to the mix

I would love to never see the phrase "little is known about" in a paper ever again. (1) Most of the time the authors are far more confident about that statement than they should be. (2) There are lots of things we know little about because no one cares. Just not knowing isn't a good motivation.

I've really got to finish writing my blog post on why topic modeling is almost never the right method to use #reviewer2

Big turnout in Olympia, WA #handsoff

I often get asked: How did I start writing? Why do I write? Who do I write for? What's my process? I procrastinated on this because, honestly, who cares about my writing process? But after repeatedly answering the same qns, I finally wrote this. eugeneyan.com/writing/writ...

Fun optical illusion. The brain still perceives it there because it expects to see it there.

What are your favorite examples of work that bridges between NLP and HCI?

Updated this post about "wishful mnemonics" in AI with notes about Phil Agre's critical technical practice and his views on "narratable" systems (from 1997!). "The purpose of AI is to build systems whose operation can be narrated using intentional vocabulary." levon003.github.io/2023/10/17/w...

Newsletter: The real threat isn't AI using open knowledge — it's AI companies killing the projects that make knowledge free. The future of free and open access isn't saying “wait, not like that” — it’s saying "yes, like that, but under fair terms”. www.citationneeded.news/free-and-ope...

SO MANY STAND UP FOR SCIENCE EVENTS TO CHOOSE FROM—153 and COUNTING! To get more information on our local events and to register your own, head to www.standupforscience2025.org/local-event-information/ ☀️⬇️🌎

The New Yorker says Wikipedia is a beacon of hope, with quotes from me and @mako.cc www.newyorker.com/news/the-led...

Important to see this type of principled stand from engineers apnews.com/article/doge...

Meta cancelling professional factchecking in favor of community notes is causing worry in West Africa, where online disinfo has led to community violence and death: globalvoices.org/2025/02/21/t... - Great Global Voices coverage.

I enjoyed this 2021 CHI paper: "Are Current Voice Interfaces Designed to Support Children’s Language Development?" dl.acm.org/doi/10.1145/... Basic point: most system's conversational scaffolding is rigid and bad, it's probably not even fair to call them "conversations"!

When I was in graduate school, nothing in the world felt more important to me than publishing a paper in the scientific literature. It was not about getting a job or competing with my peers. It was about being part of something vastly larger than myself and profoundly meaningful human history.

The CDC's political leadership is editing language that they dislike out of academic papers. The primary purpose and de facto consequence of this will be the massive reduction if not complete elimination of discussion of queer people from CDC publications. insidemedicine.substack.com/p/breaking-n...

Haven't been able to reach either Patty Murray's or Maria Cantwell's offices in the last few days. Cantwell's voice mailbox is full! I'm choosing to take that as a good sign.

I've spent the last two years trying to understand how LLMs might improve middle-school math education. I just published an article in the Journal of Educational Data Mining describing some of that work: "Designing Safe and Relevant Generative Chats for Math Learning in Intelligent Tutoring Systems"

Very good (technical) explainer answering "How has DeepSeek improved the Transformer architecture?". Aimed at readers already familiar with Transformers. epoch.ai/gradient-upd...

We cited 155 (!) papers in health, HCI, recsys, and IR for this work. One darling I refused to kill was this table I put together of the many characteristics that researchers have said might make for a good peer recommendation. Not a lot of agreement!

Can recommendation systems help people find peer support online? I'll be presenting "Peer Recommendation Interventions for Health-related Social Support: a Feasibility Assessment" at #CSCW2025! I first submitted this work in July 2022, so I'm happy to finally share this.

I highly recommend following @listifications.app just to see the totally wild set of lists you get added to. I would really love to know the inclusion criteria for some of these!