Amir Bar (@_amirbar)

Pinned post

A triple life update:
1. I recently moved from NYC to London
2. I’m starting a new lab at @imperialcollege
3. I’ve joined my friends and mentors at @amilabs as a founding Member of Technical Staff

It's time to build 🏗️🧑‍🍳

Amir Bar @_amirbar Oct 23

cool. it will be interesting to see in larger scale. Intuitively, we know that random search scales exponentially with the number of parameters whereas sgd is polynomial (for convex objectives).

spicylemonade Verified account @spicey_lemonade Oct 22

WE DID IT! We can now pretrain transformers without backprop and even beat backprop+Adam on validation loss with insane speeds.

Now the big question is scale. It’s Ironic, we have reached a breakthrough In AI capability due to inspiration from a safety blog
View quoted post on X

Amir Bar @_amirbar Oct 23

Join Jesse's tutorial if you're attending NeurIPS!

Chieh-Hsin (Jesse) Lai Verified account @JCJesseLai Oct 22

🎉 Excited to bring our book The Principles of Diffusion Models to @NeurIPSConf 2026 as a tutorial—this time, all about diffusion LLMs! 🚀

🔹 How do ideas from continuous diffusion extend to language generation?

🔹 What are the latest advances and open frontiers in diffusion LLMs?

From foundational principles to cutting-edge research. Stay tuned!

📆 Mon, Dec 7, 9 AM AEDT
📍 TikTok Ent Centre (Sydney, Australia)

w/ organizers @ssahoo_ @mittu1204 @StefanoErmon
View quoted post on X

Amir Bar @_amirbar Oct 8

NAVA-WAM: Can we learn latent actions for World Action Models?

We revisit a classic idea: jointly learning inverse and forward dynamics models while avoiding the identity shortcut (e.g., @SeonghyeonYe, @garridoq_, @jeffacce, @pulkitology, and others).

The twist: Both models are implemented within a single WAM architecture, using attention masks!

To prevent the identity shortcut, the latent actions can only sees a noisy version of the next frame, forcing the latent action to capture useful dynamics rather than simply copying the future.

This simple pretraining strategy improves both performance and efficiency across robotics benchmarks!

🧵👇

Zhaochong An @ZhaochongAn Oct 5

🤖 Abundant videos capture how the world works, but robot policies still rely on costly action labels to learn.

🔥 Excited to share our work at @Meta:

📢 Native Action-Prior Learning from Videos for World Action Models 📢

📄arxiv.org/pdf/2610.03391
🌐zhaochongan.github.io/NAVA-WAM/

🧵

View quoted post on X

Amir Bar @_amirbar Oct 6

𝗣𝗿𝗼𝗪𝗔𝗠 imagines latent visual subgoals to guide action prediction. Strong performance across robotics benchmarks and real-world experiments.
🧵👇

Fei Zhang @809428184Fei Oct 5

👋 𝗛𝗲𝗹𝗹𝗼 𝗧𝘄𝗶𝘁𝘁𝗲𝗿! 𝗙𝗶𝗿𝘀𝘁 𝘁𝘄𝗲𝗲𝘁 𝗮𝗹𝗲𝗿𝘁 🚨
Why not give it to a paper!

Excited to share our work at @Meta: 𝗣𝗿𝗼𝗪𝗔𝗠
In this paper, we try to answer:
"How should a robot imagine the future before it acts?"

🌐 Webpage: sii-ferenas.github.io/ProWAM-page/
🧵👇

View quoted post on X

Amir Bar

Latest posts 5