Yipeng Zhang

phd student @mila_quebec

[Yipeng Zhang @yipengzz]

How can we predict multiple plausible targets from a single context in joint-embedding self-supervised learning (SSL)?

Check out our paper titled “Self-Supervised Learning from Structural Invariance” accepted at #ICLR2026! Previously Best Paper Award at @unireps 2025.

https://t.co/mN5e1huPO9

We introduce AdaSSL, which models the target uncertainty and relaxes the standard assumption that the positive pair share the same semantic features.

Derived from first principles, we realize @ylecun’s JEPA with a learned latent variable for jointly learning better representations and world models, extending SSL’s utility to a broader range of data types.

1/🧵

[Xiaofeng Zhang @AZUREANg]

It’s surprisingly easy to improve intuitive physics in video generative models with a powerful latent world model ;)

Talk to @DYDYYDYYYD @adri_romsor @michal_drozdzal if you are at #CVPR

[Mengye Ren @mengyer]

Giving a talk at ICLR MemAgents Workshop tomorrow, 11:25am local, Room 205: "Does Your LLM Agent Have a Self?" Spoiler: probably not — but the reason is more nuanced than "they're just LLMs." Hope to see you there!

[Xiaofeng Zhang @AZUREANg]

I’m at #iclr2026 till Monday! DM if you’re interested in image / video generation, representation learning, physics plausibility, gen diversity, and any other.

I’ll present at
1. P4 - 3114 3-5pm Thu
2. @AIatMeta booth 9-10:30am Fri, 12-1:30pm Sat
3. Lifelong agent workshop Sun

[Xiaofeng Zhang @AZUREANg]

I’ll present our paper on how prompt complexity change the generated distribution in T2I models (https://t.co/P5x1cd6ACI) at ICLR 26!
Come and chat at Pavillon 4 - #3114 tomorrow 23 April 3PM :)

[Yipeng Zhang @yipengzz]

@hafezghm and I are in Rio presenting this work. Come by our poster and chat about SSL, world models, and continual learning!
🗓️ Saturday, April 25th, 10:30am
📍Pavilion 3-#112

[Yipeng Zhang @yipengzz]

How can we predict multiple plausible targets from a single context in joint-embedding self-supervised learning (SSL)?

Check out our paper titled “Self-Supervised Learning from Structural Invariance” accepted at #ICLR2026! Previously Best Paper Award at @unireps 2025.

https://t.co/mN5e1huPO9

We introduce AdaSSL, which models the target uncertainty and relaxes the standard assumption that the positive pair share the same semantic features.

Derived from first principles, we realize @ylecun’s JEPA with a learned latent variable for jointly learning better representations and world models, extending SSL’s utility to a broader range of data types.

1/🧵

[Divyat Mahajan @divyat09]

Presenting Future Summary Prediction at #ICLR2026 🇧🇷

📌 Friday, 24th April, 10.30 am
📍Pavilion 3 (521)

Come over to chat about novel pretraining objectives for LLMs!

[Emiliano Penaloza @emilianopp_]

I’m in Rio presenting this work at the SPOT workshop 🇧🇷☀️!

Feel free to reach out to chat about self distillation/privileged information and/or anything to do with post-training/long horizon agents.

[Lucas Maes @lucasmaes_]

JEPA are finally easy to train end-to-end without any tricks!

Excited to introduce LeWorldModel: a stable, end-to-end JEPA that learns world models directly from pixels, no heuristics.

15M params, 1 GPU, and full planning <1 second.

📑: https://t.co/cpTzgvbTS0

[Emiliano Penaloza @emilianopp_]

Come check out the talk for a deep breakdown of my recent work/blog :)

[Benjamin Thérien @ MLSys 2026 @benjamintherien]

Are frontier LLMs trained across datacenters? One thing is certain: if the pre-training optimizer’s critical batch size is too small, they are NOT! Excited to announce MuLoCo, a pre-training optimizer that can efficiently pre-train across datacenters while having large enough batch sizes to warrant doing so. 🧵1/N

[Randall Balestriero @randall_balestr]

World Modeling research needs fast iteration, reproducibility, optimized baselines, open-source, and precise zero-shot stress testing. Here comes stable-worldmodel!
Paper: https://t.co/aGmoYOId8U
Code: https://t.co/YN3PS5xxV9
Come stress-test your model/idea!
DINO-WM results ⬇️

[Emiliano Penaloza @emilianopp_]

https://t.co/SAZYxtVHcC

[Sébastien Lachapelle @seblachap]

I had a lot of fun meeting all the smart people at this workshop and presenting my work "On the Identifiability of Latent Action Policies" as an oral! A huge thanks to the organizers!

Paper: https://t.co/FqTwVTxI4Z

[Emiliano Penaloza @emilianopp_]

Remember all the self-distillation papers that came out last week. Well, we also propose it 😅, but…
But alongside something better 😎 π-Distill

We show that with this method, you can distill closed-source frontier models even tho their traces are hidden 🔒.

Both our methods can reach and even surpass the performance of the industry-standard SFT + RL with access to reasoning traces 🤯.

🔬And we spent ~100,000 hours GPU hours on a comprehensive analysis, not because the method is finicky, but because we wanted to understand why it works so well.

🧵
1/10

[World Modeling Workshop @worldmodel_conf]

What an awesome first day! Thank you all for joining and listening to our amazing speakers: @SchmidhuberAI, @sherryyangML, @cosmo_shirley, @Yoshua_Bengio, @ylecun, @mido_assran

World Models have beautiful days ahead. This is just the beginning 🫡

[Yipeng Zhang @yipengzz]

I'm at @worldmodel_26 now through Friday. Lmk if you want to chat!

[Yipeng Zhang @yipengzz]

How can we predict multiple plausible targets from a single context in joint-embedding self-supervised learning (SSL)?

Check out our paper titled “Self-Supervised Learning from Structural Invariance” accepted at #ICLR2026! Previously Best Paper Award at @unireps 2025.

https://t.co/mN5e1huPO9

We introduce AdaSSL, which models the target uncertainty and relaxes the standard assumption that the positive pair share the same semantic features.

Derived from first principles, we realize @ylecun’s JEPA with a learned latent variable for jointly learning better representations and world models, extending SSL’s utility to a broader range of data types.

1/🧵

[Yipeng Zhang @yipengzz]

We hope AdaSSL inspires new ideas for using joint-embedding SSL to learn better representations and world models on naturally structured data.

Kudos to my amazing collaborators: @hafezghm @yololulu_ @ShahabBakht @NeuralEnsemble @lcharlin

Paper: https://t.co/mN5e1huPO9. Code coming soon.

See you in Rio 🇧🇷

🧵/🧵

[Yipeng Zhang @yipengzz]

How can we predict multiple plausible targets from a single context in joint-embedding self-supervised learning (SSL)?

Check out our paper titled “Self-Supervised Learning from Structural Invariance” accepted at #ICLR2026! Previously Best Paper Award at @unireps 2025.

https://t.co/mN5e1huPO9

We introduce AdaSSL, which models the target uncertainty and relaxes the standard assumption that the positive pair share the same semantic features.

Derived from first principles, we realize @ylecun’s JEPA with a learned latent variable for jointly learning better representations and world models, extending SSL’s utility to a broader range of data types.

1/🧵

[Yipeng Zhang @yipengzz]

Across experiments, AdaSSL-V consistently improves both contrastive and distillation-based SSL. AdaSSL-S reliably improves contrastive SSL, but less so with distillation.

Why? Here are the plots of r-space usage (sparsity and diversity of the gated modules), on InfoNCE vs BYOL. With BYOL, AdaSSL-S often underutilizes r, reflected by lower diversity.

Hypothesis: (sample- or dimension-)contrastive objectives explicitly regularize information content in the embeddings, which forces the model to use r. Distillation lacks this direct pressure, so r may need extra regularization.

Curious to see how this finding affects practical use cases…

10/🧵