A short reflection on "interdisciplinary" in AI, from Lucy Suchman, @siree.sh, and myself: www.cambridge.org/core/journal...

As someone that came of age in a time where the linux distros I used had easy access to a *huge* variety of loyal clients (xmpp, so many web things), this paper and its framing are a lovely way to describe the thing I'd love for computing to return to ❤️

It's me! I am both this type of guy and, regrettably, this guy

at this point anyone who sees any actual value in these tools has to be rooting for the bubble to pop so the moneyed freaks propping it up move on to quantum computing already

Great quote from a great article, and it drives me *nuts* how big the gap is between the real affordances of AI, which can be fun and adaptive and agency affirming the political project it is inextricable from, software-brained imperialism :/

Me agreeing to a new project knowing full well it’s gonna add stress 🇲🇽

My hot take is that the median social system breaks under too much optimization pressure, and we should stop trying to optimize things

This is a great thread, and reminds me of interviewing someone at one of the big labs for a study, and they made the claim that "classification was solved in 2022" That's an insane quote, until you account for the money, time, and labor that will be poured into any problem that's deemed worth it

They need to make it so you can mass add starter packs to a list / feed if you don't want them to fill up your main feed

friendly reminder to run docker system prune if you haven't recently

Really excited about this work w/ my long-time collaborators at Boulder! We address limitations in existing morphosyntactic annotation systems for digitally under-resourced languages and show how *jointly* predicting morphological segmentation helps with glossing performance

if you invented public libraries today, every opinion page in the country would be arguing for means tested subsidized Amazon Prime memberships

It's time to seize the means of prediction

I should not learn the lesson that if I wait long enough, someone else will do it. however.

You have a moral imperative to refuse to work with these people or develop models for these purposes.

*CL folks, a recent history question: was the move to OpenReview as the review platform for *CL conferences related to the move to ARR? Or did they just happen simultaneously? Asking for a discussion with @shaily99.bsky.social

I'm really not a fan of the way some journals in the physical sciences state results upfront, and bury methods much later in the paper, or even in supplementary materials. Maybe it's my ML bias, but I don't trust that your methods are prima facie reasonable!! Show me what you did before the results!

Appointment reading

I think my AI beliefs are quickly becoming ‘this has enormous positive potential if implemented responsibly. it will not be implemented responsibly and like maybe five companies are even trying.’

This is a great thread, and makes me think about how the affordances of LLMs (largely text-only chat) preclude the kind of epistemic vigilance we were able to develop around tech like search. I think we can build LLM-based tech that's good here, but chat ain't it

Rename LASSO and Ridge to kiki and bouba

This does feel like situation where open models are held back by the state of benchmarks. For an open model, there isn't as much of a theory of use outside of benchmarks. For closed models, the companies putting them out can build for business cases and their associated internal benchmarks

Most tech workers, even at Palantir (probably), are pro-social compared to the tech owner and VC classes. They need unions so their voices matter for company governance. www.wired.com/story/palant...

Vivid mental image of cellino and barnes being like penn and teller, where one is always completely silent and if you advance a surprising legal theory your consultation is free

Firewatch is one of the most lovely games I've played. I've taken so much joy from playing it with multiple friends and seeing how different a game it is for them. I didn't know games could tell stories like this before it (It also played a small part in my fiancée and me getting together!)

This is a real banger of a paper. The example of a model being weirdly focused on jasmine (lol) makes me increasingly think that single-point-of-access models don't really consider who their audience is. Jasmine is a super legible cultural marker for people outside, but is so, _so_ generic.

“Position paper” is just a label to make some kinds of interdisciplinary theoretical work fit into the CS publishing schema.

extremely on brand, love this town

Best use of this meme i’ve ever seen

Another trip to Bangalore, another time that Google Maps (but no other Google service) turns itself Indian accented and maybe can't be turned back? 🤔

This is very millennial of me, but I really want a buzzfeed-style article summarizing the gas town discourse

This is a really interesting paper, and there's an interesting structural angle to the narrowing of focus. ML approaches involve building an infrastructure - data, models, etc. Once that infra exists, it's easy to apply it to similar problems - leading to the kind of clustering described here...

"We should keep build-and-test empiricism—its speed, openness, and compounding progress are real achievements—but we should supplement it with a second language of empiricism that lets us state and evaluate claims with more precision."

Researchers: Aside from moderation and safety features, what would you want in a researcher-focused social app? Obviously a feed of shared papers, but what else?

The two hardest problems in Computer Science are 1. Human communication 2. Getting people in tech to believe that human communication is important

This is an excellent piece on the way that a culture of repurposed and thin measurement has really warped what it means to do science. The risk of AI here is not just intensifying the use of these metrics, but reifying those flawed standards into systems that make the culture harder to change.

As a PhD student in computer science, the kind of knowledge production that is being "innovated" away here is exactly what departments like mine could use a lot more of. This is a real loss for CMU.

"the constraint is the feature" is an excellent way to put this. General purpose tools are only general purpose until you use them for something!

What if technology didn’t feel so… hollow? Some friends and I just released a manifesto about a world where tech leaves us feeling nourished (along with an evolving list of theses about how we can build it) resonantcomputing.org

A NeurIPS feed! Pin it and follow along.

It's strange to be in a field wishing for the bubble around you to pop, but I think this post captures some of the why. AI as technology has so much cool and fun potential, and it's going to be so exciting to see it freed from being developed and pushed by the awful people at the top of tech.

This is so exciting!! As someone that works on scientific PDFs, the extent to which format constraints limit how we represent the complexity of the science we do is maddening. I hope other venues adopt things like this soon!

i don't agree with the attitude of some faculty that their job is to train people who can run the code, even if they don't understand the underlying physics i want a society of people who are trained to be curious about the code they're being asked to run and capable of understanding it, actually

Please respond to this survey if you have changed or have thought about changing your name in academic publishing! For any reason, whether it be transition, recognizability, marriage, privacy, immigration, cultural reasons, etc. Please RT for reach :)

love the construction of this dataset, it feels very much of a piece with other work like the reversal curse. LLMs clearly don't "learn" or "know" in the ways we do, whether relation reversal or simple compositionality, and I love how clearly this work demonstrates that.

My name is Ozymandias, King of Kings! Look upon my works, which achieve state-of-the art performance across a diverse suite of benchmarks

✨I’m on the academic job market ✨ I’m a PhD candidate at @hcii.cmu.edu studying tech, labor, and resistance 👩🏻‍💻💪🏽💥 I research how workers and communities contest harmful sociotechnical systems and shape alternative futures through everyday resistance and collective action More info: cella.io

How to not do computational humanities: (1) Lay out a question/hypothesis about a complex, cultural domain. (2) Compute numbers that were inspired by (1) but without sufficiently formalizing (1) so as to meaningfully link to it. (3) Interpret numbers to mean whatever your prior vibes were.