Andrea Lathrop's Avatar

Andrea Lathrop

@cabernet.bsky.social

Wine... Cognitive Science... Photography... tech stuff... former Aslin Lab (Master's) and Eppler Lab (with Jackie Gibson emeritus) so I'm a weird mix of Gibsonian affordances, visual statistical learning and anticipatory eye movements. Sometimes academic.

1,365 Followers  |  2,757 Following  |  5,394 Posts  |  Joined: 01.07.2023  |  1.8241

Latest posts by cabernet.bsky.social on Bluesky

Or Mars, really, but I'm not as sentimentally attached to a pristine Mars. Though if your *real* goal is making actual human humans (eventually) 'mutiplanetary' I should think you'd start with a place humans could survive, such as terraforming Europa.

09.02.2026 00:35 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I suppose this fits with Elon's companies' tendencies to "learn faster" by iterating quickly over failures, instead of long, careful planning & flawless execution, and the Moon gives him more shots on goal than Mars. I think it's weird that one can just decide unilaterally to build on the Moon.

09.02.2026 00:35 β€” πŸ‘ 3    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I like this framing:

08.02.2026 23:49 β€” πŸ‘ 5    πŸ” 1    πŸ’¬ 0    πŸ“Œ 0
Post image

Huh...

08.02.2026 23:46 β€” πŸ‘ 4    πŸ” 1    πŸ’¬ 2    πŸ“Œ 1

To clarify, I find it absurd that embodiment and world models could be "too... anything" for Yann, but especially funny that something could be "too robot" to be real/true. He is multi-talented, and this is squarely in his field of research. (I'm not sure what the Legion of Honor has to do with it.)

08.02.2026 22:59 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I think you are talking past me, here...

08.02.2026 22:52 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

And they aren't allowed to play with robots???

08.02.2026 22:49 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

It's so cute, how could you even be mad?

08.02.2026 22:38 β€” πŸ‘ 4    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

My French is rusty, but I'm laughing at "trop robot pour Γͺtre vraiβ€Š?" (Is Yann's project too robot to be true?!?)

08.02.2026 22:34 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 2    πŸ“Œ 0

LOL.

08.02.2026 21:17 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

(Please don't be ImprovBoston... please don't be ImprovBoston...)

08.02.2026 20:33 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

(p.s. do we know which theater?) (Is this NYC? People's? Magnet? UCB?)

08.02.2026 20:32 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I am already Cog-Sci-adjacent to this mess. Please don't tell me I am also improv-adjacent. I don't need to recognize more names on uncomfortable lists.

08.02.2026 20:30 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 2    πŸ“Œ 0

It was both thrilling and immiserating, at the time, I probably have slight PTSD from it, and would absolutely do it again, in a heartbeat.

08.02.2026 20:11 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Most of the main players in the advertising audience measurement space release some kind of report every Super Bowl on which ads performed the best. We had the ambitious task of mixing all of our live data through algorithms quickly so we could report our ad leaderboard live. Quite stressful. (3/3)

08.02.2026 20:03 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

We picked a different city each year, reserved a hotel conference room, set up large screen TVs & couches, and wired people up to measure heart rate, breathing, galvanic skin response, a few other biometric data types (and a small subset were also eye tracked) while they watched the Super Bowl live.

08.02.2026 20:03 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Oh, it was my job for several years in a row circa 2012-ish. I was a Senior Analyst at a startup called Innerscope Research, which did a lot of research-for-hire in advertising and market research. My primary role was eyetracking & data analysis, but Super Bowl was an all-hands annual event. (1/3)

08.02.2026 20:03 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0
Post image

This headline in The Chronicle makes it sound like most adjuncts are independently wealthy when the reality is they have to hold second and third jobs just to make ends meet.

08.02.2026 16:44 β€” πŸ‘ 105    πŸ” 20    πŸ’¬ 7    πŸ“Œ 10

It might just be me, but this sounds suspiciously like the bros are about rediscover support vector machines.

08.02.2026 17:25 β€” πŸ‘ 2    πŸ” 1    πŸ’¬ 2    πŸ“Œ 0

Oh, that's useful... I may need to spin up some Alts.

LiveJournal had a pull-down menu to choose the audience for each post, so you could choose it on the fly. I liked that. You could sort your followers into groups, and choose which group saw the post. (ATProto breaks this externally, anyway.)

08.02.2026 14:57 β€” πŸ‘ 0    πŸ” 1    πŸ’¬ 0    πŸ“Œ 0

I’ve seen several good arguments that chat interfaces are the wrong approach to AI. I’m always persuaded while I read themβ€”and when I stop, I always wonder why history has refused to listen.

E.g this author seems right that chat interrupts flow and forces you to focus on the tool, not the code. +

08.02.2026 12:03 β€” πŸ‘ 59    πŸ” 4    πŸ’¬ 14    πŸ“Œ 0

Maybe I just need selectable alts with a single login.

08.02.2026 14:22 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

I need a social media that separates my posts into tracks, so I can post academically to academics, silly yes-and-ing among improv castmates, political stuff to my wonks, and since it's hard not to sh*tPoast sometimes, a filter so everything I said after "academics" doesn't drive away my academics.

08.02.2026 14:22 β€” πŸ‘ 3    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

the most impactful thing Elon Musk has ever done or will ever do regarding science and innovation in this country is illegally cut billions of dollars of research funds for literally no reason

08.02.2026 03:01 β€” πŸ‘ 4443    πŸ” 977    πŸ’¬ 38    πŸ“Œ 20
Preview
He Was Laid Off at The Washington Post After Working There 60 Years

AI will never replace the institutional knowledge and memories of an absolute legend who was *listening to the police scanner the night of the Watergate break-in.*

www.nytimes.com/2026/02/05/b...

08.02.2026 07:51 β€” πŸ‘ 218    πŸ” 48    πŸ’¬ 2    πŸ“Œ 1
Video thumbnail

✨ New 3D pose estimation method from my lab! #FMPose3D allows for monocular (i.e. single camera) 2D➑️3D πŸ”₯

Led by Ti Wang & w/ Xiaohang Yu #FMPose3D is SOTA on human & animal 3D benchmarks, & will be integrated into @deeplabcut.bsky.social ⬇️

πŸ“ arxiv.org/abs/2602.05755
➑️ xiu-cs.github.io/FMPose3D/

08.02.2026 07:09 β€” πŸ‘ 66    πŸ” 11    πŸ’¬ 1    πŸ“Œ 0

Idly wondering which of my marketable skills are the least replaceable by agentic AI.

08.02.2026 02:27 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

I ❀️ Keytar Bear.

08.02.2026 02:04 β€” πŸ‘ 1    πŸ” 1    πŸ’¬ 0    πŸ“Œ 0

Just saw Keytar Bear playing a Red Line stop in single digit windchill. They are one of the best parts about living in Boston. Kudos to people who make public joy in tough times.

02.02.2026 14:09 β€” πŸ‘ 24    πŸ” 8    πŸ’¬ 2    πŸ“Œ 2
Preview
Postdoctoral Research Fellow, Cognitive/Computational Neuroscience If you are a current Barnard College employee, please use the internal career site to apply for this position. Job: Postdoctoral Research Fellow, Cognitive/Computational Neuroscience The Barnard Visua...

🚨Job alert! I'm recruiting a postdoc! If you want to study the time course of task-driven visual perception, please reach out! #neuroskyence #VisionScience #CogSci barnard.wd1.myworkdayjobs.com/en-US/Facult...

07.02.2026 20:13 β€” πŸ‘ 29    πŸ” 22    πŸ’¬ 2    πŸ“Œ 0

@cabernet is following 20 prominent accounts