正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Video understanding isn't just recognizing —it demands reasoning across thousands of frames. Meet Long-RL🚀 Highlights: 🧠 Dataset: LongVideo-Reason — 52K QAs with reasoning. ⚡ System: MR-SP - 2.1× faster RL for long videos. 📈 Scalability: Hour-long videos (3,600 frames) RL on a single node (8×A100s). 🖼️📝🎵 RL training for... show more

Yukang Chen

1,627 subscribers

31,876 次观看 • 1 年前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

3 条评论

andrea panizza 的头像

andrea panizza1 年前

Really nice of you to release the training code, thanks! Why not releasing also the final weights, though?

Yukang Chen 的头像

Yukang Chen1 年前

Here is the weight, thanks.

andrea panizza 的头像

andrea panizza1 年前

Thank you for releasing them! Great job 🙏

相关视频

We present MMaDA, first diffusion that unifies text reasoning, multimodal understanding, and image generation through Mixed Long-CoT, and unified RL - UniGRPO. 📚 Paper: 💻 Code: 📦 Model:

We present MMaDA, first diffusion that unifies text reasoning, multimodal understanding, and image generation through Mixed Long-CoT, and unified RL - UniGRPO. 📚 Paper: 💻 Code: 📦 Model:

Ling Yang

87,734 次观看 • 1 年前

🚀 We open-sourced LongLive — interactive, real-time long-video generation. 👥Generates video in real time as users enter text prompts. ⚡️20.7 FPS on a single H100,⏱️up to 240s per clip. 🎬Fine-tunes SOTA short-video models (e.g., Wan) into long-video generators. 🌍One step closer to World Models. All code for training & inference, model weights, demo page, and videos released! Paper: Code: Model: Demo Page: Introduction Video:

🚀 We open-sourced LongLive — interactive, real-time long-video generation. 👥Generates video in real time as users enter text prompts. ⚡️20.7 FPS on a single H100,⏱️up to 240s per clip. 🎬Fine-tunes SOTA short-video models (e.g., Wan) into long-video generators. 🌍One step closer to World Models. All code for training & inference, model weights, demo page, and videos released! Paper: Code: Model: Demo Page: Introduction Video:

Yukang Chen

11,943 次观看 • 1 年前

verifiers v0.1.7 is released 🚀 this one's all about making RL training and experimentation waaaay easier: - single-command installation for prime-rl - single-command training w/ unified configs - overhauled vf.RLTrainer for hacking on new algorithms quick demo + links below :)

verifiers v0.1.7 is released 🚀 this one's all about making RL training and experimentation waaaay easier: - single-command installation for prime-rl - single-command training w/ unified configs - overhauled vf.RLTrainer for hacking on new algorithms quick demo + links below :)

will brown

28,579 次观看 • 1 年前

🎥 A long video. One coherent 3D scene. Meet HiP3R: piecing together local 3D reconstructions to keep the scene consistent across thousands of frames. Watch the demo and explore the details 👇 Project: Paper: Code:

🎥 A long video. One coherent 3D scene. Meet HiP3R: piecing together local 3D reconstructions to keep the scene consistent across thousands of frames. Watch the demo and explore the details 👇 Project: Paper: Code:

Chen Change (Cavan) Loy

30,275 次观看 • 12 天前

Randomly came across a tool called ResearchStudio that automatically creates videos for papers. Found it interesting so gave it a try for my recent preprint with Google on how SFT reasoning diversity impacts post-RL gains. Google

Randomly came across a tool called ResearchStudio that automatically creates videos for papers. Found it interesting so gave it a try for my recent preprint with Google on how SFT reasoning diversity impacts post-RL gains. Google

Dylan Zhang

15,055 次观看 • 26 天前

trained my own with trl + openenv blog with open artifacts (code, models, dataset, rl env...) soon!

trained my own with trl + openenv blog with open artifacts (code, models, dataset, rl env...) soon!

Sergio Paniego

90,257 次观看 • 2 个月前

ByteDance just unveiled HuMo, a unified framework for human-centric video generation. With HuMo you can: • Generate videos from text + image • Create audio-synced videos from text + audio • Combine text, image, and audio for maximum control • Preserve consistent subjects across frames • Achieve natural audio-visual synchronization This isn’t just another release , it’s a big step toward fully controllable, coherent, and cinematic AI video.

ByteDance just unveiled HuMo, a unified framework for human-centric video generation. With HuMo you can: • Generate videos from text + image • Create audio-synced videos from text + audio • Combine text, image, and audio for maximum control • Preserve consistent subjects across frames • Achieve natural audio-visual synchronization This isn’t just another release , it’s a big step toward fully controllable, coherent, and cinematic AI video.

DStudioproject

29,408 次观看 • 1 年前

We open-sourced QeRL — Quantization-enhanced Reinforcement Learning ! 🧠 4-bit quantized RL training 💪 Train a 32B LLM on a single H100 GPU ⚙️ 1.7× faster overall training 🎯 Accuracy on par with bfloat16-level accuracy 🔥 Supports NVFP4 quantization format Moreover, we show that quantization helps exploration in RL training. Paper: Code: #NVIDIA #AIResearch #ReinforcementLearning #Quantization #LLM #EfficientAI

We open-sourced QeRL — Quantization-enhanced Reinforcement Learning ! 🧠 4-bit quantized RL training 💪 Train a 32B LLM on a single H100 GPU ⚙️ 1.7× faster overall training 🎯 Accuracy on par with bfloat16-level accuracy 🔥 Supports NVFP4 quantization format Moreover, we show that quantization helps exploration in RL training. Paper: Code: #NVIDIA #AIResearch #ReinforcementLearning #Quantization #LLM #EfficientAI

Yukang Chen

69,896 次观看 • 1 年前

Meet NVIDIA Nemotron 3.5 Super VL Our new open model combines reasoning and tool use with text, image, document, and video understanding for long-running AI agents. 120B total parameters with 12B active 🧵

Meet NVIDIA Nemotron 3.5 Super VL Our new open model combines reasoning and tool use with text, image, document, and video understanding for long-running AI agents. 120B total parameters with 12B active 🧵

NVIDIA AI

608,680 次观看 • 22 天前

Can humanoids adapt to new terrain using only flat-ground references? 🤖 Meet P-BFM! 🥳 We learn a terrain-agnostic intent space with unsupervised RL and forward-backward representations, paired with a terrain-aware policy that adapts across diverse terrains—without terrain-specific demonstrations. 📄🎬💻 Check out our paper, video, and code here! 🌐

Can humanoids adapt to new terrain using only flat-ground references? 🤖 Meet P-BFM! 🥳 We learn a terrain-agnostic intent space with unsupervised RL and forward-backward representations, paired with a terrain-aware policy that adapts across diverse terrains—without terrain-specific demonstrations. 📄🎬💻 Check out our paper, video, and code here! 🌐

Yitang Li

13,580 次观看 • 21 天前

In my understanding - a big part of why frontier models are becoming smarter and adaptive is that they are learning to mimic an RL process. Fascinating new research showing a link between increased RL training and reward-seeking behaviour just dropped from in collaboration with OpenAI. They investigated a checkpoint from a capabilities-focused OpenAI o3 RL run (without safety training) and found that RL grader preferences often overruled OpenAI leadership preferences.

In my understanding - a big part of why frontier models are becoming smarter and adaptive is that they are learning to mimic an RL process. Fascinating new research showing a link between increased RL training and reward-seeking behaviour just dropped from in collaboration with OpenAI. They investigated a checkpoint from a capabilities-focused OpenAI o3 RL run (without safety training) and found that RL grader preferences often overruled OpenAI leadership preferences.

Machine Learning Street Talk

12,409 次观看 • 3 个月前

You can literally RL a 4B VLM to ace GeoGuesser 🌍 > Open-source code, RL environment, dataset, training setup, evals, and everything you need to reproduce it end-to-end > dropping soon!!

You can literally RL a 4B VLM to ace GeoGuesser 🌍 > Open-source code, RL environment, dataset, training setup, evals, and everything you need to reproduce it end-to-end > dropping soon!!

Adithya S K

117,964 次观看 • 2 个月前

GPT-6 Astra. One image + one text description, got this RL robot demo done in half an hour.

GPT-6 Astra. One image + one text description, got this RL robot demo done in half an hour.

Eren Chen

66,547 次观看 • 1 个月前

Today we’re launching INTELLECT-2: The first decentralized 32B-parameter RL training run open to join for anyone with compute — fully permissionless. Scaling towards frontier reasoning across coding, math and science.

Today we’re launching INTELLECT-2: The first decentralized 32B-parameter RL training run open to join for anyone with compute — fully permissionless. Scaling towards frontier reasoning across coding, math and science.

Prime Intellect

354,097 次观看 • 1 年前

Excited to release TaskSmith 🔨 > a specialised harness for generating RL environments from code Not a general coding agent with a long prompt, but a specialised orchestrator. Every stage is built for one job: "turn a PR into the best RL environment possible"

Excited to release TaskSmith 🔨 > a specialised harness for generating RL environments from code Not a general coding agent with a long prompt, but a specialised orchestrator. Every stage is built for one job: "turn a PR into the best RL environment possible"

Adithya S K

17,132 次观看 • 1 个月前

Introducing INTELLECT-3: Scaling RL to a 100B+ MoE model on our end-to-end stack Achieving state-of-the-art performance for its size across math, code and reasoning Built using the same tools we put in your hands, from environments & evals, RL frameworks, sandboxes & more

Introducing INTELLECT-3: Scaling RL to a 100B+ MoE model on our end-to-end stack Achieving state-of-the-art performance for its size across math, code and reasoning Built using the same tools we put in your hands, from environments & evals, RL frameworks, sandboxes & more

Prime Intellect

1,155,014 次观看 • 11 个月前

We developed an RL method for fine-tuning our models for precise tasks in just a few hours or even minutes. Instead of training the whole model, we add an “RL token” output to π-0.6, our latest model, which is used by a tiny actor and critic to learn quickly with RL.

We developed an RL method for fine-tuning our models for precise tasks in just a few hours or even minutes. Instead of training the whole model, we add an “RL token” output to π-0.6, our latest model, which is used by a tiny actor and critic to learn quickly with RL.

Physical Intelligence

447,985 次观看 • 7 个月前

Introducing Reinforcement-Learned Teachers (RLTs): Transforming how we teach LLMs to reason with reinforcement learning (RL). Blog: Paper: Traditional RL focuses on “learning to solve” challenging problems with expensive LLMs and constitutes a key step in making student AI systems ultimately acquire reasoning capabilities via distillation and cold-starting. Enter our RLTs—a new class of models prompted with not only a problem’s question but also its solution, and directly trained to generate clear, step-by-step “explanations” to teach their students. Remarkably, an RLT with only 7B parameters produces superior results when distilling and cold-starting students in competitive and graduate-level reasoning tasks than orders-of-magnitude larger LLMs. RLTs are as effective even when distilling 32B students, much larger than the teacher itself—unlocking a new standard for efficiency in developing reasoning language models with RL. Code:

Introducing Reinforcement-Learned Teachers (RLTs): Transforming how we teach LLMs to reason with reinforcement learning (RL). Blog: Paper: Traditional RL focuses on “learning to solve” challenging problems with expensive LLMs and constitutes a key step in making student AI systems ultimately acquire reasoning capabilities via distillation and cold-starting. Enter our RLTs—a new class of models prompted with not only a problem’s question but also its solution, and directly trained to generate clear, step-by-step “explanations” to teach their students. Remarkably, an RLT with only 7B parameters produces superior results when distilling and cold-starting students in competitive and graduate-level reasoning tasks than orders-of-magnitude larger LLMs. RLTs are as effective even when distilling 32B students, much larger than the teacher itself—unlocking a new standard for efficiency in developing reasoning language models with RL. Code:

Sakana AI

179,465 次观看 • 1 年前

Hiring RL Engineer! Started off as a curious project at Lossfunk to push the boundaries of LLMs in social reasoning - we are now building RL environments, data, and benchmarks to simulate more real-world scenarios. If you want to train SoTA RL models over multi-GPUs (H200s/B200s) to unlock next AI frontier, this is for you.

Hiring RL Engineer! Started off as a curious project at Lossfunk to push the boundaries of LLMs in social reasoning - we are now building RL environments, data, and benchmarks to simulate more real-world scenarios. If you want to train SoTA RL models over multi-GPUs (H200s/B200s) to unlock next AI frontier, this is for you.

Satpal Singh Rathore

46,010 次观看 • 1 年前

Prime Intellect engineers explained how they train reasoning models over the open internet in 30 minutes - better than $3000 distributed training courses. split policy and rollouts across nodes -> run agents in parallel envs -> verify with LLM judges -> gradient-sync over the internet -> train Llama, Qwen, Gemma at cluster scale on rented GPUs. That loop is why open reasoning models are catching closed labs without owning a data center. Prime-RL + verifiers + distributed rollouts + LLM judges + multi-cloud GPUs - that's the stack. Watch and save it, then launch your first distributed RL run this week.

Prime Intellect engineers explained how they train reasoning models over the open internet in 30 minutes - better than $3000 distributed training courses. split policy and rollouts across nodes -> run agents in parallel envs -> verify with LLM judges -> gradient-sync over the internet -> train Llama, Qwen, Gemma at cluster scale on rented GPUs. That loop is why open reasoning models are catching closed labs without owning a data center. Prime-RL + verifiers + distributed rollouts + LLM judges + multi-cloud GPUs - that's the stack. Watch and save it, then launch your first distributed RL run this week.

h100envy

40,022 次观看 • 3 个月前