Xiao-Xiong [email protected]

Theoretical systems neuroscience at Color-blind super-scientist Mary trying to navigate the career path. Looking for a job.

[Xiao-Xiong [email protected] @lin_xiaoxiong]

My first neuroscience work is available online! check it out! 👉
https://t.co/4JfABWRhUA
w\ Andreas Nieder & @lab_jacob

👇 a rundown of how we link individual neurons🧠 to representational geometry 🌐 through the perspective of neuronal implementation✨

1/n

[Xiao-Xiong [email protected] @lin_xiaoxiong]

@walterzhu8 how much credit did these 1.5 days burn?

[Xiao-Xiong [email protected] @lin_xiaoxiong]

@thsottiaux 8 hours after I used my reset card🥲

[Jürgen Schmidhuber @SchmidhuberAI]

@PopBase https://t.co/StC4rvRcZJ

[Kyros @IamKyros69]

Before you ask AI another dumb coding question… watch this.

[Xiao-Xiong [email protected] @lin_xiaoxiong]

The Roses of Heliogabalus irl

[Xiao-Xiong [email protected] @lin_xiaoxiong]

finally seeing a DOS Robot Odyssey come back

[dhruv @hail_dhruv]

forget leetcode, I'm farming food with python.

[Joscha Bach @Plinz]

AI is a wormhole into a new dimension of creativity

[soli @solisolsoli]

Wassily Kandinsky's Dance Curves, 1926

[Delip Rao @ COLM26 @deliprao]

Agents training in RL gym

[Yann LeCun @ylecun]

Hey Europe, you want a vibrant tech industry, right?
Do this immediately (it's genius!)
https://t.co/h48w0fTKDJ

[Xiao-Xiong [email protected] @lin_xiaoxiong]

Neuroscientists' wet dream?

[Rudy Gilman @rgilman33]

This layer in DINO-v2 dedicates about half its attention mass to a single operation. Each of the sixteen heads independently learns the same circuit to perform this task. What is this all-important operation?

The “no-op”. That’s right, we’re spending half our computation to do… absolutely nothing.

[Tankred Saanum @TankredSaanum]

Can LLMs do reinforcement learning in-context - and if so, how do they do it? Using Sparse Autoencoders, we find that Llama 3 relies on representations resembling TD errors, Q-values and even the SR to learn in three RL tasks in-context! Co-lead with the inimitable @can_demircann

[Jim Fan @DrJimFan]

@kenneth0stanley Yes, RL can drive open-ended novelty search. It’s finally time to make self-driven, sustained discovery a reality

[Xiao-Xiong [email protected] @lin_xiaoxiong]

Nice paper! I have actually wanted to do this for years...
https://t.co/VWZ2vYf6eN

[Jay Van Bavel, PhD @jayvanbavel]

One of the reason I prefer Bluesky to X (as well as Threads/Facebook/LinkedIN) is simply because my newsfeed isn't controlled by an engagement-driven algorithm

This reduces the incentives for gaming the algorithm with clickbait, conflict entreprenuers, etc. https://t.co/wWRXWxvTSx

[Xiao-Xiong [email protected] @lin_xiaoxiong]

In this video the /ch/ in German word "China" is clearly pronounced [ç] (or a bit like [x] in the male's voice), while almost all the Germans I've met in real life say [ʃ] or [k], basically "Schina" or "Kina"... has time changed so much?

[Historic Vids @historyinmemes]

Germany released this Chinese food ad in 1988

[BROPOTKIN™️ by blackhart 🐈‍⬛ @peterbropotkin]

Bro is cookin

[Xiao-Xiong [email protected] @lin_xiaoxiong]

also using Noh theater gait helps...

[Xiao-Xiong [email protected] @lin_xiaoxiong]

Proud to say my motor control system spontaneously found this solution... perhaps implicitly inspired by bike suspension

[Fermat's Library @fermatslibrary]

In 2015, Jiwon Han published a paper on why walking with a coffee mug often leads to spills.

Han discovered that our natural walking frequency unfortunately matches the frequency at which coffee prefers to slosh around.

He also explored various methods to prevent spills, including walking backwards and the "claw-hand" grip.