Lightnews — Scholar-powered news

The Weirdly Small AI That Cracks Reasoning Puzzles [HRM]

2 13

Jia-Bin Huang @jbhuang0604.bsky.social · Aug 15

New video!

A quick dive into the recent Hierarchical Reasoning Model (HRM) through the lens of algorithm synthesis.

Check it out: youtu.be/RK7lysjz_G0

1 14

Jia-Bin Huang @jbhuang0604.bsky.social · Aug 8

Diffusion LLMs are promising ways to overcome the limitations of autoregressive LLMs.

Less error propagation, easier to control, and faster to sample!

But how do Diffusion LLMs actually work? 🤔

In this video, let's explore some ideas on this fascinating topic! youtu.be/8BTOoc0yDVA

1 11

Jia-Bin Huang @jbhuang0604.bsky.social · Jul 16

In an era of billion-parameter models everywhere, it's incredibly refreshing to see how a fundamental question can be formulated and solved with simple, beautiful math.

- How should we orient a solar panel ☀️🔋? -

Zero AI! If you enjoy math, you'll love this!

Video: www.youtube.com/watch?v=ZKzL...

1 2 9

Jia-Bin Huang @jbhuang0604.bsky.social · Jul 8

*Slides without slide titles*

When I first tried presenting WITHOUT slide titles, everything flowed so much better! (totally validated ... by me)!

Give it a shot! Once you try it, you’ll never want to go back.

3

Jia-Bin Huang @jbhuang0604.bsky.social · Jul 8

*Empty initial slides*

What’s a better starting point than that default slide layout?

A completely blank slide.

It helps you explore the design space and focus on delivering a clear, compelling story.

1 1

Jia-Bin Huang @jbhuang0604.bsky.social · Jul 8

*Bullet points*

The second thing the layout prompts you to do?
("Click to add text").

Start a bullet list.

Among so many creative forms of presenting your ideas, it nudges you toward the most boring one: a list. 🔢

Jia-Bin Huang @jbhuang0604.bsky.social · Jul 8

*Slide title*

The first thing this layout does is to ask you to add a slide title.

Seems reasonable, right? visuals, this encourages you to
1) lead your presentation with text instead of visuals and
2) cram in many titles in a talk, making it harder to maintain a narrative flow.

Jia-Bin Huang @jbhuang0604.bsky.social · Jul 8

Why is the "Title and Content" slide layout BAD?

Most people prepare their presentation from this default layout. I used it for years without questioning it.

BUT, this essentially guides you toward developing poor presentation. Why? 🤔

5 2 22

Jia-Bin Huang @jbhuang0604.bsky.social · Jul 2

Thanks! Yup, I hope to cover some fun computer vision applications. Stay tuned!

Policy Gradient in One Minute

Jia-Bin Huang @jbhuang0604.bsky.social · Jul 1

Kids’ summer camp just kicked off, and that means...
I finally have time to make new videos!

What topics are you most interested in right now?

1 5

Jia-Bin Huang @jbhuang0604.bsky.social · Jun 24

Why More Researchers Should be Content Creators

Just trying something new! I recorded one of my recent talks, sharing what I learned from starting as a small content creator.

youtu.be/0W_7tJtGcMI

We all benefit when there are more content creators!

1 2 8

Reposted by Jia-Bin Huang

Kosta Derpanis @csprofkgd.bsky.social · Jun 19

Fresh out of the oven! 🍞 @jbhuang0604.bsky.social breaks down Mean Flow from Kaiming’s group in his latest video.

Video: youtu.be/swKdn-qT47Q?...

2 18

Jia-Bin Huang @jbhuang0604.bsky.social · Jun 20

No time? I’ve got your back!

Check out Policy Gradient in One Minute!
youtu.be/p9k9YUdnNlk

Have fun!

One Step, Big Leap: The Simple Idea Transforming Generative AI

4

Jia-Bin Huang @jbhuang0604.bsky.social · Jun 20

Policy gradient methods rock!

These are the core techniques for making your transformer "chat" and "reason", a robot that manipulates objects, and a drone that maneuvers in a complex environment.

BUT, how do we learn all the developments in the past 30+ years?

1 1 3

Jia-Bin Huang @jbhuang0604.bsky.social · Jun 20

Check out the video to learn this new, elegant formulation of generative models!

youtu.be/swKdn-qT47Q

1 13

Jia-Bin Huang @jbhuang0604.bsky.social · Jun 20

Awesome! 🤩

So glad to hear the authors enjoyed the video, totally made my day!

1 13

Jia-Bin Huang @jbhuang0604.bsky.social · Jun 17

We had a blast at CVPR2025!

There was so much to learn! I am particularly excited to meet many new friends and reconnect with old ones.

I feel energized. Already looking forward to the next one!

6

Jia-Bin Huang @jbhuang0604.bsky.social · Jun 4

Thanks a lot!

Jia-Bin Huang @jbhuang0604.bsky.social · Jun 4

Kullback–Leibler (KL) divergence is a cornerstone of machine learning.

We use it everywhere, from training classifiers and distilling knowledge from models, to learning generative models and aligning LLMs.

BUT, what does it mean, and how do we (actually) compute it?

Video: youtu.be/tXE23653JrU

1 5 31

Jia-Bin Huang @jbhuang0604.bsky.social · Jun 3

My X/Twitter account has been hacked... Please don't believe what they said!

Trying to get it back in the meantime. Sorry for the inconvenience!

5

Jia-Bin Huang @jbhuang0604.bsky.social · May 21

How LLMs Learn to Reason with Reinforcement Learning

Full video: www.youtube.com/watch?v=mg-i...

3

Jia-Bin Huang @jbhuang0604.bsky.social · May 21

Ha! Yes, Seungjae insisted that we call this IVE.