Just finished my poster for #MPSA this year, entirely in Quarto/Typst! 9.1/10 experience after seeing how fast it compiled compared to my Beamer poster last year :D
Maybe less stylish, but I have full faith someone like @andrew.heiss.phd can grind out a neat template in the future haha
Really great paper, kinda funny that the UC system doesn't grant access to an article in Nature on research transparency
🧵A thread on "AI one-shot papers".
This has not been a focus on Bluesky, but on Twitter/X and LinkedIn, there has been a large discussion of the ability to "one-shot" policy evaluation, which David Yanagizawa-Drott has been pursuing in an interesting way: ape.socialcatalystlab.org
This entire thing gives off Oakland Wye vibes
New version of our paper (w/ @tzkli.bsky.social) on estimating pre-trends placebos in regression imputation DiD!
3 big changes!
A classmate who went to SF for the Dem convention said that from her POV all of the candidates were basically saying the same thing with slightly different wording. Wondering if this is just another example of that or if the pollster memo spec might explain it
Chromebooks were fun until my school district blocked the dinosaur game on them
no wonder chalamet hyped up marty supreme so much
That was probably the founder's reaction when she realized her son was screwing over the company in the 2010s
Imagine taking something that's led to tons of death threats against professionals in different sports (video games included) and thinking it would be a good idea to stake it on politicians' lives instead
Instead of just partisan cues and social identities, Americans can base their vote on whoever will maximize their Kalshi returns
Would be curious to see if a prompt like this works since I've had ChatGPT try to create BibTeX citations based on an article's DOI and it would consistently spew out incorrect entries while claiming it matched the DOI to the "official [journal name] publisher metadata."
"...how'd it end up like this?"
"John Roberts has made his decision, now let him..."
I just reopened Zen for the first time in probably a year - definitely impressed to see how much it's changed since download. Might switch from Arc if they've brought tab-sync over now!
ACS data strong
2025 has sure been one hell of a year for publishing mishaps
This is terrifying.
"[AI agents] can... infer a researcher's latent hypotheses and produce data that artificially confirms them."
...
"We can no longer trust that survey responses are coming from real people" -@seanjwestwood.bsky.social
Ehh, I used to get a lot of very entertaining Instagram reels where users would ask Grok leading questions (you can probably guess the type of user who would) only to have Grok give out the answer they weren't expecting... I'm guessing those reels aren't going to pop up on my feed anymore.
Surprised it took them this long honestly
(for the record i'm just pointing out the reference to grimmer as an observation to the op, no bad intentions were inferred)
Done!!!! Hopefully @hadley.nz finds them funny ahaha
Funny enough it does cite Grimmer and Stewart (2013), then calls it a day
Update: looks like they fixed it with a new release!
Currently writing up a six_seven() and thick_of_it() function as we speak - stay tuned
Sad to see yet another lawmaker - Rep. Jared Golden in this instance - citing threats against their loved ones as an impediment to their continued public service.
New, from me: DHS says immigration enforcement employees like ICE are facing unprecedented threats and 1000% increase in assaults. But the data does not add up. Look at the details, and we see state agents using excessive force, and then lying about it.
donmoynihan.substack.com/p/whos-threa...
Apparently there's a dash between "causal" and "inference." The preprint gets it correct but the website... somehow omits the dash.
Explaining the Defying Gravity x God Is transition to one of my professors was not on my bucket list for 2025 but it's happened somehow
