Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

FP4 Explore, BF16 Train Diffusion Reinforcement Learning via Efficient Rollout Scaling paper:

AK

499,672 subscribers

12,761 views • 1 month ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Physics3D: Learning Physical Properties of 3D Gaussians via Video Diffusion Paper: (not yet) Project: Code:

Physics3D: Learning Physical Properties of 3D Gaussians via Video Diffusion Paper: (not yet) Project: Code:

MrNeRF

19,907 views • 2 years ago

Frontier models that use reasoning to "think" during inference are generating 5X more AI tokens per year. ✨ "Inference is now a thinking process. And in order to teach AI how to think, reinforcement learning and very significant computation was introduced into post-training," said NVIDIA CEO Jensen Huang in a recent keynote at #CES2026. Reinforcement learning is increasing computation demands across all AI scaling laws: pre-training, post-training, and test-time scaling. Learn more about reinforcement learning and AI scaling ➡️

Frontier models that use reasoning to "think" during inference are generating 5X more AI tokens per year. ✨ "Inference is now a thinking process. And in order to teach AI how to think, reinforcement learning and very significant computation was introduced into post-training," said NVIDIA CEO Jensen Huang in a recent keynote at #CES2026. Reinforcement learning is increasing computation demands across all AI scaling laws: pre-training, post-training, and test-time scaling. Learn more about reinforcement learning and AI scaling ➡️

NVIDIA Data Center

13,159 views • 4 months ago

🤔Want a principled way to RL your diffusion model? Check Data-regularized Reinforcement Learning (DDRL)! Post-train NVIDIA #Cosmos World Foundation models with a million GPU hours! 🤯 Novel formulation ➡️ Theoretically integrates SFT into RL ➡️ Robust to Reward Hacking 🛑 Details: #DDRL #Diffusion #RL #NVIDIA #Cosmos

🤔Want a principled way to RL your diffusion model? Check Data-regularized Reinforcement Learning (DDRL)! Post-train NVIDIA #Cosmos World Foundation models with a million GPU hours! 🤯 Novel formulation ➡️ Theoretically integrates SFT into RL ➡️ Robust to Reward Hacking 🛑 Details: #DDRL #Diffusion #RL #NVIDIA #Cosmos

Haotian Ye

77,456 views • 6 months ago

Reinforcement learning

Reinforcement learning

Massimo

3,407,828 views • 2 years ago

ManipTrans is out on Hugging Face Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning

ManipTrans is out on Hugging Face Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning

AK

22,136 views • 1 year ago

Redwood AI | Mobility Reinforcement Learning

Redwood AI | Mobility Reinforcement Learning

1X

1,317,736 views • 1 year ago

Our CoRL 2024 paper shows Reinforcement Learning can allow robots to learn skills via real-world practice, without any demonstrations or simulation engineering. Rewards are provided using language/vision models, and mobility of robots enables autonomous exploration. 1/N

Our CoRL 2024 paper shows Reinforcement Learning can allow robots to learn skills via real-world practice, without any demonstrations or simulation engineering. Rewards are provided using language/vision models, and mobility of robots enables autonomous exploration. 1/N

Russell Mendonca

38,454 views • 1 year ago

MatAnyone 2 is out on Hugging Face Scaling Video Matting via a Learned Quality Evaluator paper: app:

MatAnyone 2 is out on Hugging Face Scaling Video Matting via a Learned Quality Evaluator paper: app:

AK

34,753 views • 3 months ago

An interesting example of reinforcement learning.

An interesting example of reinforcement learning.

Historic Vids

1,563,912 views • 3 years ago

Watch Gemini 2.5 Pro implement a landmark Google DeepMind research paper. 🕹️ It codes the reinforcement learning algorithm, visualizes the training live and even debugs errors. ↓

Watch Gemini 2.5 Pro implement a landmark Google DeepMind research paper. 🕹️ It codes the reinforcement learning algorithm, visualizes the training live and even debugs errors. ↓

Google DeepMind

369,872 views • 1 year ago

An Interesting example of Reinforcement learning!! 🐓

An Interesting example of Reinforcement learning!! 🐓

Aether

7,058,785 views • 2 years ago

Nvidia presents Fast-dLLM v2 Efficient Block-Diffusion LLM

Nvidia presents Fast-dLLM v2 Efficient Block-Diffusion LLM

AK

18,753 views • 8 months ago

This is a single uncut video, showing a robot learning several tasks instantly, after just one demonstration each ... This is possible because we've now been able to achieve in-context learning for everyday robotics tasks, and I'm very excited to announce our latest paper: 🎆 Instant Policy: In-Context Imitation Learning via Graph Diffusion 🎆 (1/6) 🧵👇

This is a single uncut video, showing a robot learning several tasks instantly, after just one demonstration each ... This is possible because we've now been able to achieve in-context learning for everyday robotics tasks, and I'm very excited to announce our latest paper: 🎆 Instant Policy: In-Context Imitation Learning via Graph Diffusion 🎆 (1/6) 🧵👇

Edward Johns

74,663 views • 1 year ago

Getting Cyn ready for reinforcement learning >:3

Getting Cyn ready for reinforcement learning >:3

DT

14,256 views • 7 months ago

NeurIPS 2025 Paper: LLMs are Reinforcement Learners 🤯! Surprisingly, we show that LLMs can solve RL tasks without any external component! We introduce Prompted Policy Search (ProPS), an RL method based only LLMs and in-context learning. [Paper]

NeurIPS 2025 Paper: LLMs are Reinforcement Learners 🤯! Surprisingly, we show that LLMs can solve RL tasks without any external component! We introduce Prompted Policy Search (ProPS), an RL method based only LLMs and in-context learning. [Paper]

Heni Ben Amor

51,172 views • 6 months ago

notes on my reinforcement learning for rocket league side quest so far: i’m in awe that — we live in a time where i can train a reinforcement learning model that is learning to play rocket league and at the same time, test community made rocket league models. where, both processes can run in parallel & locally on my hardware. in the game window — i’m playing against a reinforcement learning ppo model that’s running locally. and in the vscode window — i’m training my own reinforcement learning model (just a dummy example script for now). also, the rocket league botting ecosystem has incredible tooling where you can test out different reinforcement learning ideas against other algorithms + it has great documentation for linux and windows. it’s honestly a beautiful hidden gem of a rabbit hole :)

notes on my reinforcement learning for rocket league side quest so far: i’m in awe that — we live in a time where i can train a reinforcement learning model that is learning to play rocket league and at the same time, test community made rocket league models. where, both processes can run in parallel & locally on my hardware. in the game window — i’m playing against a reinforcement learning ppo model that’s running locally. and in the vscode window — i’m training my own reinforcement learning model (just a dummy example script for now). also, the rocket league botting ecosystem has incredible tooling where you can test out different reinforcement learning ideas against other algorithms + it has great documentation for linux and windows. it’s honestly a beautiful hidden gem of a rabbit hole :)

naklecha

46,302 views • 1 year ago

Baby steps towards physically based animation with reinforcement learning...

Baby steps towards physically based animation with reinforcement learning...

Dennis Gustafsson

32,132 views • 1 year ago

(1/5) FP4 hardware is here, but 4-bit attention still kills model quality, blocking true end-to-end FP4 serving. To fix that, we propose Attn-QAT, the first systematic study of quantization-aware training for attention. The result: FP4 attention quality is comparable to BF16 attention with 1.1x–1.5x higher throughput than SageAttention3 on an RTX 5090 and 1.39x speedup over FlashAttention-4 on a B200. Blog: Code: Checkpoints:

(1/5) FP4 hardware is here, but 4-bit attention still kills model quality, blocking true end-to-end FP4 serving. To fix that, we propose Attn-QAT, the first systematic study of quantization-aware training for attention. The result: FP4 attention quality is comparable to BF16 attention with 1.1x–1.5x higher throughput than SageAttention3 on an RTX 5090 and 1.39x speedup over FlashAttention-4 on a B200. Blog: Code: Checkpoints:

Hao AI Lab

37,175 views • 1 month ago

RL-100 Performant Robotic Manipulation with Real-World Reinforcement Learning

RL-100 Performant Robotic Manipulation with Real-World Reinforcement Learning

AK

15,364 views • 7 months ago

Daniel Kokotajlo says companies are making AI more agentic by scaling RL (reinforcement learning) and connecting agents to various tools. by early 2027, these agents will be fully autonomous, superhuman coders able to replace human programmers.

Daniel Kokotajlo says companies are making AI more agentic by scaling RL (reinforcement learning) and connecting agents to various tools. by early 2027, these agents will be fully autonomous, superhuman coders able to replace human programmers.

Haider.

11,559 views • 1 year ago