jessicahullman.bsky.social - Profile | ThreadSky | a Reddit-style client for Bluesky

Why are we sometimes skeptical about the prospect of LLMs producing "interesting" writing? statmodeling.stat.columbia.edu/2025/02/25/w...

submitted 1 day ago • 1 comment

Wrote a blog post on how when I want to reflect on the state of my heart, I just think about what it would mean to truly believe in perfectly zero effects or completely stable processes 🙂 statmodeling.stat.columbia.edu/2025/02/20/p...

submitted 6 days ago • 0 comments

Who is teaching LLM/GenAI for social science research? Could you share your syllabus with me? I'm trying to create a reading group for a small group of friends and colleagues.

submitted 13 days ago • 2 comments

PhD students in social sciences studying "metascience" and AI: This looks like an AMAZING opportunity. Fully funded research postdoc that can be housed at whatever university you can find an appropriate mentor. sloan.org/programs/dig...

submitted 13 days ago • 1 comment

What are prediction sets good for? It turns out just as calibration is the "right" way of quantifying uncertainty for risk-neutral (expectation maximizing) decision makers, prediction sets are the right way of quantifying uncertainty for certain kinds of risk-averse decision makers.

submitted 20 days ago • 2 comments

I’ve learned its worth stopping what I’m doing to read danah’s posts when they come out and this one is no exception

submitted 23 days ago • 0 comments

Many of us are aghast at the unprecedented dismantlement of the US administrative state. All around me, people are trying to connect what’s happening to historical events. Here's my attempt to offer a few possible frames for sense-making right now: www.zephoria.org/thoughts/arc...

submitted 23 days ago • 8 comments

RIP Marianne Faithfull m.youtube.com/watch?v=GXrJ...

submitted 25 days ago • 0 comments

If this administration reminds of nothing else, it makes clear something I say all the time: race is the central organizing feature of American politics and American life. Nothing else comes close. These fools can’t even talk about a tragic plane crash without talking about “DEI.”

submitted 27 days ago • 3 comments

Okay, where do stand now on the accuracy of imprecise credences? Still living with the impossibility w.r.t strictly proper scoring rules? @wiglet1981.bsky.social @kevinzollman.com @jeffhelzner.bsky.social @jessicahullman.bsky.social

submitted 28 days ago • 3 comments

Some sad but important truths in this blogpost. We should all read it "It’s almost as if questioning the premises or foundations of certain research threatens the very “personhood” of the researchers involved."

submitted 34 days ago • 0 comments

@economeager.bsky.social's excellent essay got me thinking about how we confuse research with affirming (& preserving) our identity. Where we can tolerate uncertainty but only to the extent it blows in the direction of our unfolding personal narrative statmodeling.stat.columbia.edu/2025/01/23/s...

submitted 34 days ago • 2 comments

New starter pack: go.bsky.app/8g64qRL

submitted 38 days ago • 8 comments

Sometimes you have to write “Truth and Probability” just because Margaret won’t fuck with you. open.substack.com/pub/rottenan...

submitted 40 days ago • 21 comments

Consider submitting to our upcoming CHI workshop on augmenting human reasoning with AI!

submitted 40 days ago • 0 comments

Was talking to my 7yo abt how girls mature faster, but she didn’t get “on average”, so I’m like “Imagine randomly picking a girl & boy, & each time seeing who’s more mature. More often it will be the girl” Then realize I learned something too! “And this is why we call it probability of superiority”

submitted 40 days ago • 0 comments

Is it possible to establish consensus in interpreting the probability of an individual event? How far does exchangeability get us toward reconciling frequentist vs personalist (eg Bayesian) views? Summarized part of Dawid's "On Individual Risk" here: statmodeling.stat.columbia.edu/2025/01/17/h...

submitted 40 days ago • 1 comment

This piece with quotes from @profsanderlinden.bsky.social and @jayvanbavel.bsky.social does a bit of a dive into whether fact-checking works, surrounding the Meta platform policy changes. It notes a large (N=20K) meta-analysis indicates a significant positive effect. www.nature.com/articles/d41...

submitted 44 days ago • 2 comments

I don't know, maybe this is just me, but, it might help to remove the pejorative association with this kind of work by engaging with these people in good faith, acknowledging the fruits of their efforts, and treating them as collaborators rather than adversaries.

submitted 46 days ago • 6 comments

An interesting thread on updating an intro CS course to encourage good use of LLMs.

submitted 51 days ago • 0 comments

This term I'll be teaching a graduate course on Human-AI Interaction. Check out the reading list and let me know if I am missing any of your favorites: docs.google.com/document/d/1... #HAI #Human-AI #HumanComputerInteraction

submitted 54 days ago • 7 comments

Progress in 2024 (Jessica) statmodeling.stat.columbia.edu/2025/01/02/p...

submitted 55 days ago • 0 comments

Just like it was “recently shown” that people actually have varying preferences

submitted 57 days ago • 0 comments

Part III: the symbol grounding problem, iconic-indexical semantics, limitations and possibilities of language, Laozi's anti-technocratic skepticism, Zhuangzi's pluralism and sense of wonder, the dao of layered architectures, constraints that deconstrain. realizable.substack.com/p/the-daoist...

submitted 58 days ago • 2 comments

Need fresh ideas for difficult holiday conversations? Try convincing someone who believes that aiming for calibrated predictions is always enough that they are missing valuable second order uncertainty statmodeling.stat.columbia.edu/2024/12/31/c... Happy new year!

submitted 57 days ago • 0 comments

I don’t have time for round two, but you should compare the paper to the prereg for this one. www.nature.com/articles/s41...

submitted 65 days ago • 3 comments

Next time you feel yourself compelled to mention statistical guarantees, talk about statistical oomph instead. Same vibes, less letters. Personally I’m holding out for “With a sinking heart, our method learns a tenuous mapping on past data we hope is related to future data.”

submitted 68 days ago • 3 comments

From 1/1/24 to 11/5/24 we tracked every article featured in top 20 slots on homepages of 10 major publishers. Using a combination of human oversight & AI categorization, we assign a category, topic, sub-topic to every article. Here is how much they talked about Biden's Age v. Trump's Project 2025:

submitted 69 days ago • 13 comments

Enhance!

submitted 71 days ago • 2 comments

In case its helpful for junior female academics, a strategy I often use when I suspect I'm getting asked to do service bc I'm female is to Suggest-A-Man. Safest to suggest someone w/roughly same seniority as you. Doesn't hurt to throw in a "They seem to have ideas on [topic of service]." 1/2

submitted 70 days ago • 3 comments

We're working hard behind the scenes to finalize the dates and venue for #FAccT2025! While final confirmation is still pending, our tentative conference dates are June 23-26. Expect more updates soon!

submitted 70 days ago • 0 comments

New post on The 100% CI! We unravel a series of mysterious research misconduct cases that can be traced back to a single villain striking from the shadow: Rogue RA www.the100.ci/2024/12/18/r...

submitted 70 days ago • 2 comments

Inspired by @wimlds.bsky.social , I looked for a "Women in Machine Learning" starter pack and couldn't find one. So I created one! May have some mistakes. I'll try to do an AI ethics one next. 🤗 go.bsky.app/LT6CwNN

submitted 78 days ago • 56 comments

Come work with @jessicahullman.bsky.social and I as a postdoc! We'll apply approaches from HCI, AI/ML, statistics, and decision theory to design new tools & methods for evaluating human-AI decision-making (e.g., tools to elicit, represent, & validate specifications of real-world decision problems).

submitted 78 days ago • 1 comment