Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

training for RL 2027 🥰😋

jessica__

17,388 subscribers

27,523 görüntüleme • 1 ay önce •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Peaking with *NEW* RL 2027 Video Settings

Peaking with NEW RL 2027 Video Settings

UrNm Gyro. 🇦🇲

26,506 görüntüleme • 1 ay önce

Does LLM RL post-training need to be on-policy?

Does LLM RL post-training need to be on-policy?

Kianté Brantley

114,480 görüntüleme • 8 ay önce

New project! Flow Policy Gradients for Robot Control tldr; a simple online RL recipe for training and fine-tuning flow policies for robots co-led w/ Hongsuk Benjamin Choi:

New project! Flow Policy Gradients for Robot Control tldr; a simple online RL recipe for training and fine-tuning flow policies for robots co-led w/ Hongsuk Benjamin Choi:

Brent Yi

94,426 görüntüleme • 9 ay önce

I've gotten a mujoco sim RL training loop for a unitree robot at 200k SPS. I'm looking into the physics for friction, contact dynamics. My goal: can I reproduce & beat the mujoco playground RL baselines This is running in my web browser with raylib. Its the baseline

I've gotten a mujoco sim RL training loop for a unitree robot at 200k SPS. I'm looking into the physics for friction, contact dynamics. My goal: can I reproduce & beat the mujoco playground RL baselines This is running in my web browser with raylib. Its the baseline

kache

30,204 görüntüleme • 4 ay önce

1/ RL for improving task success has made LLMs remarkably capable. But what happens when the model fails? Turns out, training for graceful failure can be surprisingly tricky! Very excited to share our new paper: Training Language Models to Fail Gracefully 🧵

1/ RL for improving task success has made LLMs remarkably capable. But what happens when the model fails? Turns out, training for graceful failure can be surprisingly tricky! Very excited to share our new paper: Training Language Models to Fail Gracefully 🧵

Daman Arora

59,926 görüntüleme • 20 gün önce

What is Spider girl training for?🌚😋

What is Spider girl training for?🌚😋

Cleverly 💐

275,328 görüntüleme • 1 yıl önce

Why am I working on RL for LLMs, when I can go work on RL for Olaf... #humanoidssummit

Why am I working on RL for LLMs, when I can go work on RL for Olaf... #humanoidssummit

Shane Gu

12,503 görüntüleme • 10 ay önce

An electric horse you say? Dave's trying to bring me to his lair (100% RL, sim2real 60m of training)

An electric horse you say? Dave's trying to bring me to his lair (100% RL, sim2real 60m of training)

Harry O'Connor

19,078 görüntüleme • 1 yıl önce

Training hard for what’s next, I gotta look good 😋

Training hard for what’s next, I gotta look good 😋

Matthew Riddle

225,012 görüntüleme • 2 yıl önce

Introducing CQN: Coarse-to-fine Q-Network, a value-based RL algorithm for continuous control🦾Initialized with 20~50 demonstrations, it learns to solve real-world robotic tasks within 10 mins of training, without any pre-training and shaped rewards! (1/4)

Introducing CQN: Coarse-to-fine Q-Network, a value-based RL algorithm for continuous control🦾Initialized with 20~50 demonstrations, it learns to solve real-world robotic tasks within 10 mins of training, without any pre-training and shaped rewards! (1/4)

Younggyo Seo

16,438 görüntüleme • 2 yıl önce

GPT-6 Astra test 4/n Prompt: "Implement pen spinning with a dexterous hand. Use Isaac Lab for RL training, use the Sharpa hand, and create the pen mesh yourself. Give me a trained RL policy and a visualization video. You are free to search the web and download papers or anything else you need." After running autonomously for a day and a half (including training the policy), this is what we got: Credit: my student Chengyang Li

GPT-6 Astra test 4/n Prompt: "Implement pen spinning with a dexterous hand. Use Isaac Lab for RL training, use the Sharpa hand, and create the pen mesh yourself. Give me a trained RL policy and a visualization video. You are free to search the web and download papers or anything else you need." After running autonomously for a day and a half (including training the policy), this is what we got: Credit: my student Chengyang Li

Wentao Zhu

695,884 görüntüleme • 1 ay önce

Reinforcement learning should be able to improve upon behaviors seen when training. In practice, RL agents often struggle to generalize to new long-horizon behaviors. Our new paper studies *horizon generalization*, the degree RL algorithms generalize to reaching distant goals. 1/

Reinforcement learning should be able to improve upon behaviors seen when training. In practice, RL agents often struggle to generalize to new long-horizon behaviors. Our new paper studies horizon generalization, the degree RL algorithms generalize to reaching distant goals. 1/

Vivek Myers

79,523 görüntüleme • 1 yıl önce

🚨 New: Integrating Harbor (Harbor Framework) for end-to-end Computer-Use evaluation(for Windows and Linux) at scale with Thinking Machines' Tinker, OSWorld, Daytona, and bare-metal servers. We just added support for Computer Use, Tinker, and OSWorld to Harbor - a framework for evaluating agents and generating RL training data by running large-scale rollouts across parallel sandboxed environments and collecting trajectories for SFT and RL. Repo and blogpost below 👇

🚨 New: Integrating Harbor (Harbor Framework) for end-to-end Computer-Use evaluation(for Windows and Linux) at scale with Thinking Machines' Tinker, OSWorld, Daytona, and bare-metal servers. We just added support for Computer Use, Tinker, and OSWorld to Harbor - a framework for evaluating agents and generating RL training data by running large-scale rollouts across parallel sandboxed environments and collecting trajectories for SFT and RL. Repo and blogpost below 👇

Marco Mascorro

19,448 görüntüleme • 8 ay önce

Training large models with RL typically runs on a fixed homogeneous cluster of the latest accelerators for numerical stability and faster throughput. This throws away inference optimization work done for other vendors and generations, limits the capacity you can utilize, and is unnecessarily expensive. So we trained a model on agentic coding tasks using fully asynchronous RL and mixed-precision, mixed-accelerator samplers. With off-policy correction, training stayed stable through convergence, on par with homogeneous samplers, at a lower cost.

Training large models with RL typically runs on a fixed homogeneous cluster of the latest accelerators for numerical stability and faster throughput. This throws away inference optimization work done for other vendors and generations, limits the capacity you can utilize, and is unnecessarily expensive. So we trained a model on agentic coding tasks using fully asynchronous RL and mixed-precision, mixed-accelerator samplers. With off-policy correction, training stayed stable through convergence, on par with homogeneous samplers, at a lower cost.

Modal

167,236 görüntüleme • 22 gün önce

fell for a g*rl in the holy month….

Sensitive content

fell for a g*rl in the holy month….

lalo

163,942 görüntüleme • 8 ay önce

here’s some dopamine for your wednesday afternoon 😋🥰

here’s some dopamine for your wednesday afternoon 😋🥰

G00N0NYM0US

13,058 görüntüleme • 1 yıl önce

RL is painfully slow 😭 — bottlenecked by super-long CoT rollout. 🔭 Sparse attention should help, but naive sparse rollout hits a brutal efficiency–stability tradeoff: A tedious trial-and-error sparsity sweep for each dense policy is required before an actual RL run. 🐤Sparrow chirps no more pain! Introduce Sparrow: Sparse Rollout for stable and efficient long-context RL. Sparrow finds that: 💡As long as we keep the tail distribution mismatch throughout the sparse rollout above a critical threshold, the RL training will be stable. 💡Even cooler! Through comprehensive control studies of Qwen3-1.7B, 4B, 8B thinking models RL with 40K rollout max length, the critical threshold stays constant across model sizes. 💡Sparrow then finds the optimal dynamic sparse schedule to reach the threshold with minimal cost. 💡Sparrow's findings are empirically validated to generalize in Qwen3-14B, and hold on both Math and Coding RL. 🐤Sparrow empirically helps achieve 2.2× / 2.4× / 2.0× rollout speedup on Qwen3 1.7B / 4B / 8B thinking models, while keeping training stability over extended RL steps. We release the 🐤bird in the following formats. [1/n] Paper: Code: Blog:

RL is painfully slow 😭 — bottlenecked by super-long CoT rollout. 🔭 Sparse attention should help, but naive sparse rollout hits a brutal efficiency–stability tradeoff: A tedious trial-and-error sparsity sweep for each dense policy is required before an actual RL run. 🐤Sparrow chirps no more pain! Introduce Sparrow: Sparse Rollout for stable and efficient long-context RL. Sparrow finds that: 💡As long as we keep the tail distribution mismatch throughout the sparse rollout above a critical threshold, the RL training will be stable. 💡Even cooler! Through comprehensive control studies of Qwen3-1.7B, 4B, 8B thinking models RL with 40K rollout max length, the critical threshold stays constant across model sizes. 💡Sparrow then finds the optimal dynamic sparse schedule to reach the threshold with minimal cost. 💡Sparrow's findings are empirically validated to generalize in Qwen3-14B, and hold on both Math and Coding RL. 🐤Sparrow empirically helps achieve 2.2× / 2.4× / 2.0× rollout speedup on Qwen3 1.7B / 4B / 8B thinking models, while keeping training stability over extended RL steps. We release the 🐤bird in the following formats. [1/n] Paper: Code: Blog:

Infini-AI-Lab

79,496 görüntüleme • 4 ay önce