Video yükleniyor...

Video Yüklenemedi

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

Ana Sayfaya Dön

Thanks AK! Finally, robot can do continuous, agile, autonomous, adaptive jumping over stair and stepping stone Key idea: combine the pros of model-free RL and model-based control. RL (for CoM refs) + QP (for GRF) + WBC (for torque) Open-sourced:

Guanya Shi

7,637 subscribers

32,155 görüntüleme • 1 yıl önce •via X (Twitter)

Bilim & Teknoloji

Anya Rossi• Live Now

Private livecam show

4 Yorum

Allen Z. Ren profil fotoğrafı

Allen Z. Ren1 yıl önce

@_akhaliq It reminds me I was trying to get the Minitaur jumping the stairs (scripted controller) somewhere in Hamerschlag I think in 2018...Amazing to see how both software and hardware have gotten so much better since then!

Yaoyao(Freax) Qian profil fotoğrafı

Yaoyao(Freax) Qian1 yıl önce

@_akhaliq Interesting work!

Kashu Yamazaki profil fotoğrafı

Kashu Yamazaki1 yıl önce

@_akhaliq This is impressive!

Heng (Alfredo)Zhang profil fotoğrafı

Heng (Alfredo)Zhang1 yıl önce

@_akhaliq very impressive!

Benzer Videolar

Google has open-sourced GNM Head, its parametric model for generating 3D human heads The model, based on real-world scan data, provides 250 controls for head shape, and 380+ for facial expressions A free add-on integrates it into Blender #B3D

Google has open-sourced GNM Head, its parametric model for generating 3D human heads The model, based on real-world scan data, provides 250 controls for head shape, and 380+ for facial expressions A free add-on integrates it into Blender #B3D

CG Channel

45,976 görüntüleme • 1 ay önce

GPT-5.5 by Reasoning Effort: I've asked it in Codex to create a physics-based visualisation of RL cycles for different sized models (70b, 1t, 10t), to demonstrate how the amount of RL you can do differs by model size. My assessment of each: - Low: weird slop - Medium: kinda cooked - High: sort of tried but ultimately incoherent - Extra High: elite - really nice idea and well executed Obviously this is just one shot, but worth trying different reasoning levels for the new models, medium seems to be pretty good for GPT-5.5 and it was really bad for many previous GPT models.

GPT-5.5 by Reasoning Effort: I've asked it in Codex to create a physics-based visualisation of RL cycles for different sized models (70b, 1t, 10t), to demonstrate how the amount of RL you can do differs by model size. My assessment of each: - Low: weird slop - Medium: kinda cooked - High: sort of tried but ultimately incoherent - Extra High: elite - really nice idea and well executed Obviously this is just one shot, but worth trying different reasoning levels for the new models, medium seems to be pretty good for GPT-5.5 and it was really bad for many previous GPT models.

Peter Gostev (SF: 22-26 June)

209,258 görüntüleme • 3 ay önce

MolmoAct2 is landing in LeRobot! Ai2's open Action Reasoning Model combines a Molmo2-ER vision-language backbone with a flow-matching continuous action expert to predict robot action chunks from images, language instructions, and proprioceptive state. An open robot foundation model built for real-world control, with strong out-of-the-box performance and easy fine-tuning in LeRobot. Pick-and-place inference running on NVIDIA DGX Spark! Blog: Paper: Thanks to Ai2 Jiafei Duan Haoquan Fang

MolmoAct2 is landing in LeRobot! Ai2's open Action Reasoning Model combines a Molmo2-ER vision-language backbone with a flow-matching continuous action expert to predict robot action chunks from images, language instructions, and proprioceptive state. An open robot foundation model built for real-world control, with strong out-of-the-box performance and easy fine-tuning in LeRobot. Pick-and-place inference running on NVIDIA DGX Spark! Blog: Paper: Thanks to Ai2 Jiafei Duan Haoquan Fang

LeRobot

24,727 görüntüleme • 2 ay önce

Introducing RL Environment Creator Skill Now any one can create RL environments $ npx skills add adithya-s-k/RL_Envs_101 > You can create environments across multiple frameworks like OpenEnv, OpenReward, Verifiers, NemoGym ... > the repo has live working examples of environments that your coding agent can reference > The skill is design to first understand what type of model you are training and create an environment while keeping that in mind ps. There’s a lot more to building RL environments that can be used for training. One major aspect is the data, which this skill can’t directly solve. However, the skill will help with implementing tools, rewards, and other components of an RL environment, making it easier to go from idea to implementation quickly across different frameworks. Let me know if you’d be interested in a detailed, end-to-end blog/tutorial on building an environment and actually training a model for a useful use case.

Introducing RL Environment Creator Skill Now any one can create RL environments $ npx skills add adithya-s-k/RL_Envs_101 > You can create environments across multiple frameworks like OpenEnv, OpenReward, Verifiers, NemoGym ... > the repo has live working examples of environments that your coding agent can reference > The skill is design to first understand what type of model you are training and create an environment while keeping that in mind ps. There’s a lot more to building RL environments that can be used for training. One major aspect is the data, which this skill can’t directly solve. However, the skill will help with implementing tools, rewards, and other components of an RL environment, making it easier to go from idea to implementation quickly across different frameworks. Let me know if you’d be interested in a detailed, end-to-end blog/tutorial on building an environment and actually training a model for a useful use case.

Adithya S K

46,556 görüntüleme • 3 ay önce

AI agents can now control robots! For ClawCon, we integrated OpenClaw🦞 and Robot Operating System (ROS) - the largest open-source robotics stack powering millions of robots worldwide. If you had an autonomous agent IRL, what would you make it do?

AI agents can now control robots! For ClawCon, we integrated OpenClaw🦞 and Robot Operating System (ROS) - the largest open-source robotics stack powering millions of robots worldwide. If you had an autonomous agent IRL, what would you make it do?

Vitaly Bulatov

153,129 görüntüleme • 6 ay 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

78,717 görüntüleme • 2 ay önce

System identification (sysid) is the process of finding the physical parameters that make a simulation match reality. If you're training an RL locomotion policy in simulation, the accuracy of your motor model directly affects how well the policy transfers to the real robot. A recent git commit by Kevin Zakka added a sysid toolbox to MuJoCo which automates this process: you provide recorded motor data and a MuJoCo model, and it optimizes the model parameters to minimize the difference between simulated and real trajectories. For my RobStride Dynamics RS02 QDD motors (17 Nm peak, 7.75:1 gear), I built a Rust tool that sends multi-sine torque excitation at 1 kHz and records position/velocity feedback. I then feed this data into MuJoCo's sysid optimizer.

System identification (sysid) is the process of finding the physical parameters that make a simulation match reality. If you're training an RL locomotion policy in simulation, the accuracy of your motor model directly affects how well the policy transfers to the real robot. A recent git commit by Kevin Zakka added a sysid toolbox to MuJoCo which automates this process: you provide recorded motor data and a MuJoCo model, and it optimizes the model parameters to minimize the difference between simulated and real trajectories. For my RobStride Dynamics RS02 QDD motors (17 Nm peak, 7.75:1 gear), I built a Rust tool that sends multi-sine torque excitation at 1 kHz and records position/velocity feedback. I then feed this data into MuJoCo's sysid optimizer.

David Bar

48,347 görüntüleme • 4 ay önce

Model-Free Reinforcement Learning (MFRL) has been alluring, especially with supercharged compute with physics on GPU. However, the methods use 0-th order gradients, and are often not the best optimizers. Can we do better than PPO in continuous control for robotics? Turns out yes! 🥳 tl;dr: Faster, better RL than PPO in continuous control 💪 The answer lies in using more information from the simulation. We are juicing the simulation on GPU as it is, why not use it for gradients as well? This has been a driving question in a series of our works. We first studied this problem in ICLR 2022 paper on Short Horizon Actor Critic Naive gradient based methods are stuck in local minima and have exploding/vanishing gradients. SHAC solved this problem truncated rollouts and model based value estimation, where the model is Differentiable Sim. This boosted sample efficiency and wall-clock time immensely especially in high dimensional systems such as humanoids Yet, given enough compute PPO often caught up. Our follow up paper on on Adaptive Horizon Actor Critic at ICML 2024 discovers the cause and provides a fix. However, we find that even when given ground-truth dynamics, not all gradients are useful due to sample error. 1st-Order Model-Based Reinforcement Learning methods employing differentiable simulation provide gradients with reduced variance but are susceptible to bias in scenarios involving stiff dynamics, such as physical contact. We find that back-propagating through contact and long trajectories drastically reduces gradient accuracy. Using this insight, we propose AHAC to dynamically adapt its roll-out horizon to avoid differentiating through stiff contact. AHAC is a first-order model-based RL algorithm that learns high-dimensional tasks in minutes (wall clock) and outperforms PPO by 40%, even in the limit of data provided to PPO. This work is led by Ignat Georgiev alongside Krishnan Srinivasan, Jie Xu, Eric Heiden and ample assistance from warp team at NVIDIA Robotics (Miles Macklin)

Model-Free Reinforcement Learning (MFRL) has been alluring, especially with supercharged compute with physics on GPU. However, the methods use 0-th order gradients, and are often not the best optimizers. Can we do better than PPO in continuous control for robotics? Turns out yes! 🥳 tl;dr: Faster, better RL than PPO in continuous control 💪 The answer lies in using more information from the simulation. We are juicing the simulation on GPU as it is, why not use it for gradients as well? This has been a driving question in a series of our works. We first studied this problem in ICLR 2022 paper on Short Horizon Actor Critic Naive gradient based methods are stuck in local minima and have exploding/vanishing gradients. SHAC solved this problem truncated rollouts and model based value estimation, where the model is Differentiable Sim. This boosted sample efficiency and wall-clock time immensely especially in high dimensional systems such as humanoids Yet, given enough compute PPO often caught up. Our follow up paper on on Adaptive Horizon Actor Critic at ICML 2024 discovers the cause and provides a fix. However, we find that even when given ground-truth dynamics, not all gradients are useful due to sample error. 1st-Order Model-Based Reinforcement Learning methods employing differentiable simulation provide gradients with reduced variance but are susceptible to bias in scenarios involving stiff dynamics, such as physical contact. We find that back-propagating through contact and long trajectories drastically reduces gradient accuracy. Using this insight, we propose AHAC to dynamically adapt its roll-out horizon to avoid differentiating through stiff contact. AHAC is a first-order model-based RL algorithm that learns high-dimensional tasks in minutes (wall clock) and outperforms PPO by 40%, even in the limit of data provided to PPO. This work is led by Ignat Georgiev alongside Krishnan Srinivasan, Jie Xu, Eric Heiden and ample assistance from warp team at NVIDIA Robotics (Miles Macklin)

Animesh Garg

52,308 görüntüleme • 2 yıl önce

hi all, excited to join! i'm building an expressive mini "shoggoth" robot which will eventually be hooked up to gpt4o realtime voice. i'm currently working on the low-level policies, which are trained in a mujoco simulation with RL. to delay working on raw-pixels for now, i trained a pose-estimation model using deeplabcut and triangulate the position in 3d space using the stereo cameras. eventually, i'll use gpt4o's tool calling capabilities to activate several of these policies (closed and open loop) based on the dialog flow! captions: manual actuation of the tentacle / 3d pose estimation / target design

hi all, excited to join! i'm building an expressive mini "shoggoth" robot which will eventually be hooked up to gpt4o realtime voice. i'm currently working on the low-level policies, which are trained in a mujoco simulation with RL. to delay working on raw-pixels for now, i trained a pose-estimation model using deeplabcut and triangulate the position in 3d space using the stereo cameras. eventually, i'll use gpt4o's tool calling capabilities to activate several of these policies (closed and open loop) based on the dialog flow! captions: manual actuation of the tentacle / 3d pose estimation / target design

Matthieu LC

45,587 görüntüleme • 1 yıl önce

🚀New paper out - We present Video-MSG (Multimodal Sketch Guidance), a novel planning-based training-free guidance method for T2V models, improving control of spatial layout and object trajectories. 🔧 Key idea: • Generate a Video Sketch — a spatio-temporal plan with background, foreground, and motion in the pixel space. • Encode this structure directly into the latent space of the diffusion model during generation, which does not require fine-tuning or additional memory during inference. 🧵

🚀New paper out - We present Video-MSG (Multimodal Sketch Guidance), a novel planning-based training-free guidance method for T2V models, improving control of spatial layout and object trajectories. 🔧 Key idea: • Generate a Video Sketch — a spatio-temporal plan with background, foreground, and motion in the pixel space. • Encode this structure directly into the latent space of the diffusion model during generation, which does not require fine-tuning or additional memory during inference. 🧵

Jialu Li

35,060 görüntüleme • 1 yıl önce

With Hunyuan3D World Model 1.0 now released and open-sourced, we're excited to showcase the technical highlights behind this impressive innovation: ✅360° Panoramic Generation: Creates complete, immersive “world scenes”, far beyond localized views. ✅Explorable 3D Scene Generation: Generates diverse, spatially consistent 3D worlds from text/image for truly immersive exploration. ✅Interactive/Editable: Achieves separation of foreground objects, background terrain, ground, and sky, for seamless secondary editing. ✅Exportable Mesh: Generated scenes can be exported as 3D meshes for direct import into mainstream game engines and modeling software. ✅Industry-Leading SOTA Evaluation: Surpasses state-of-the-art open-source models in generation quality. As the industry's first open-source model for physical simulation and explorable world generation, Hunyuan3D World Model 1.0 aims to foster a collaborative community ecosystem with developers and enthusiasts. ✨ Try it now: 🤗 Hugging Face:

With Hunyuan3D World Model 1.0 now released and open-sourced, we're excited to showcase the technical highlights behind this impressive innovation: ✅360° Panoramic Generation: Creates complete, immersive “world scenes”, far beyond localized views. ✅Explorable 3D Scene Generation: Generates diverse, spatially consistent 3D worlds from text/image for truly immersive exploration. ✅Interactive/Editable: Achieves separation of foreground objects, background terrain, ground, and sky, for seamless secondary editing. ✅Exportable Mesh: Generated scenes can be exported as 3D meshes for direct import into mainstream game engines and modeling software. ✅Industry-Leading SOTA Evaluation: Surpasses state-of-the-art open-source models in generation quality. As the industry's first open-source model for physical simulation and explorable world generation, Hunyuan3D World Model 1.0 aims to foster a collaborative community ecosystem with developers and enthusiasts. ✨ Try it now: 🤗 Hugging Face:

Tencent Hy

23,203 görüntüleme • 1 yıl önce

Big moment for text-to-speech. Qwen just open-sourced a text-to-speech model that lets you clone voices, design new ones, and control speech using natural language. Let me explain what I mean: You can literally tell it "speak in a cheerful tone with slight nervousness," and it actually does that. No complex audio engineering needed. What makes this special: - 3-second voice cloning - Covers 10 languages: English, German, French, and more - Latency as low as 97ms for real-time applications - Supports both streaming and non-streaming generation The model comes in two sizes (0.6B and 1.7B parameters), so you can pick based on your hardware and quality needs. Three modes to work with: 1. Custom Voice: Use pre-built premium voices with instruction-based style control 2. Voice Design: Describe the voice you want in plain English (or Chinese), and the model creates it 3. Voice Clone: Provide a 3-second reference audio and clone that voice The best part? It integrates with vLLM for production deployment and has a simple Python package you can pip install. I've shared a link to the GitHub repo in the next tweet.

Big moment for text-to-speech. Qwen just open-sourced a text-to-speech model that lets you clone voices, design new ones, and control speech using natural language. Let me explain what I mean: You can literally tell it "speak in a cheerful tone with slight nervousness," and it actually does that. No complex audio engineering needed. What makes this special: - 3-second voice cloning - Covers 10 languages: English, German, French, and more - Latency as low as 97ms for real-time applications - Supports both streaming and non-streaming generation The model comes in two sizes (0.6B and 1.7B parameters), so you can pick based on your hardware and quality needs. Three modes to work with: 1. Custom Voice: Use pre-built premium voices with instruction-based style control 2. Voice Design: Describe the voice you want in plain English (or Chinese), and the model creates it 3. Voice Clone: Provide a 3-second reference audio and clone that voice The best part? It integrates with vLLM for production deployment and has a simple Python package you can pip install. I've shared a link to the GitHub repo in the next tweet.

Akshay 🚀

31,249 görüntüleme • 6 ay önce

AI video is moving from one-off experiments to repeatable production workflows. Dreamina Seedance 2.5 is coming soon to the Dreamina platform. For creative and marketing teams, the upcoming model will bring: 1. Up to 50 multimodal references Combine images, videos, audio, and other creative directions in one generation. 2. Up to 30 seconds of continuous video Create longer scenes with more room for product stories and brand narratives. 3. White-model and green-screen references Improve control over movement, composition, and visual direction. 4. Multilingual creation Adapt content for more global markets. 5. Precise local editing Refine specific parts without recreating the entire video. The result: a more flexible path from creative brief to finished content. 👉 #dreamina #dreaminapartner #seedance #dreaminaseedance25

codemarch

78,823 görüntüleme • 25 gün önce

LEONARDO, also called LEO, was built by researchers at Caltech’s Center for Autonomous Systems and Technologies. Its full name means LEgs ONboARD drOne. The idea is simple but unusual: • Build a small biped robot • Give it drone-style thrust • Use the legs for ground contact • Use the propellers for balance and lift • Combine walking, hopping and flying in one system LEO is basically a hybrid between a walking robot and a flying drone. How it was built: • Two lightweight legs • Three actuated joints in each leg • Four propeller thrusters near the shoulders • A lightweight body • Leg motors for ground movement • Propellers for balance, lift and aerial control • Real-time control software that synchronizes the legs and propellers How it walks: • The legs move the robot forward • The feet touch the ground like a normal biped • The propellers constantly correct balance from above • The robot can stay upright even in unstable situations • The thrust reduces the risk of falling during difficult motions How it flies: • The legs stop being the main locomotion system • The four propellers generate lift • The robot behaves more like a drone • It can take off, fly over obstacles and land back on its legs What makes it different: • It does not walk like a normal humanoid • It does not fly like a normal drone • It blends both systems • The legs handle contact with the ground • The propellers act like fast stabilizers • The control system decides how much help comes from the legs and how much comes from thrust That is why LEO can: • Walk • Hop • Fly over obstacles • Ride a skateboard • Balance on a slackline The key idea is walking with aerial stabilization.

LEONARDO, also called LEO, was built by researchers at Caltech’s Center for Autonomous Systems and Technologies. Its full name means LEgs ONboARD drOne. The idea is simple but unusual: • Build a small biped robot • Give it drone-style thrust • Use the legs for ground contact • Use the propellers for balance and lift • Combine walking, hopping and flying in one system LEO is basically a hybrid between a walking robot and a flying drone. How it was built: • Two lightweight legs • Three actuated joints in each leg • Four propeller thrusters near the shoulders • A lightweight body • Leg motors for ground movement • Propellers for balance, lift and aerial control • Real-time control software that synchronizes the legs and propellers How it walks: • The legs move the robot forward • The feet touch the ground like a normal biped • The propellers constantly correct balance from above • The robot can stay upright even in unstable situations • The thrust reduces the risk of falling during difficult motions How it flies: • The legs stop being the main locomotion system • The four propellers generate lift • The robot behaves more like a drone • It can take off, fly over obstacles and land back on its legs What makes it different: • It does not walk like a normal humanoid • It does not fly like a normal drone • It blends both systems • The legs handle contact with the ground • The propellers act like fast stabilizers • The control system decides how much help comes from the legs and how much comes from thrust That is why LEO can: • Walk • Hop • Fly over obstacles • Ride a skateboard • Balance on a slackline The key idea is walking with aerial stabilization.

Techniahqrobot | humanoid robots

135,515 görüntüleme • 1 ay önce

And finally, my 3D model of @Lord_Griselda's Medusa is done! This was a lot of fun to do, I've actually had it finished for a couple of weeks, but I also had to finish the video for it, where I walk through the entire process which you can find below! #blender #b3d

And finally, my 3D model of @Lord_Griselda's Medusa is done! This was a lot of fun to do, I've actually had it finished for a couple of weeks, but I also had to finish the video for it, where I walk through the entire process which you can find below! #blender #b3d

Niall

100,304 görüntüleme • 2 yıl önce

AI has had exactly two scaling axes that worked so far, and the second one is starting to look finite too the first one was pretraining: with scaling parameters and data, we got world knowledge (i.e. ChatGPT had read enough to know things), but it started saturating a while ago the second one was RL, and people had been doing RL the whole time before that: RLHF is RL but it never scaled far because it was trying to control the exact output, which tokens come out, how the text reads, but you can only push that so far before you’re just polishing RLVR dropped that constraint: giving the model a task, then checking whether the final answer is right, and ignoring everything in between -- so the model does whatever it wants in the middle and only the endpoint gets graded, and that’s much closer to actual RL and it’s what bought us planning and reasoning (arguably, tool use sits around 2.5 on this list -- while useful, it's not a different kind of thing) so one axis gave knowledge, the other gave reasoning, and both of them are one model working alone the next axis is how many models you can get working on the same problem, which is a different kind of axis than the previous two we know that multi-agent RL has always been the harder problem: I spent years in that literature and the gap between single-agent and multi-agent is definitely not incremental -- it’s a whole different class of difficulty! which is also why the derivatives are steep at the start, nobody has picked the easy wins yet... and the thing that gates this multi-agent coordination is communication: models can only coordinate as well as they can exchange information, and right now they do that by writing sentences to each other imagine what could we possibly achieve if we properly open that third axis development by letting models to exchange information in their native "language" without loosing any computational data that they produce during inference

AI has had exactly two scaling axes that worked so far, and the second one is starting to look finite too the first one was pretraining: with scaling parameters and data, we got world knowledge (i.e. ChatGPT had read enough to know things), but it started saturating a while ago the second one was RL, and people had been doing RL the whole time before that: RLHF is RL but it never scaled far because it was trying to control the exact output, which tokens come out, how the text reads, but you can only push that so far before you’re just polishing RLVR dropped that constraint: giving the model a task, then checking whether the final answer is right, and ignoring everything in between -- so the model does whatever it wants in the middle and only the endpoint gets graded, and that’s much closer to actual RL and it’s what bought us planning and reasoning (arguably, tool use sits around 2.5 on this list -- while useful, it's not a different kind of thing) so one axis gave knowledge, the other gave reasoning, and both of them are one model working alone the next axis is how many models you can get working on the same problem, which is a different kind of axis than the previous two we know that multi-agent RL has always been the harder problem: I spent years in that literature and the gap between single-agent and multi-agent is definitely not incremental -- it’s a whole different class of difficulty! which is also why the derivatives are steep at the start, nobody has picked the easy wins yet... and the thing that gates this multi-agent coordination is communication: models can only coordinate as well as they can exchange information, and right now they do that by writing sentences to each other imagine what could we possibly achieve if we properly open that third axis development by letting models to exchange information in their native "language" without loosing any computational data that they produce during inference

Sasha Malysheva

11,393 görüntüleme • 10 gün önce

NVIDIA DROPPED A MOTION DIFFUSION MODEL FOR HUMANOID ROBOTS trained on 700 hours of mocap data kimodo generates high-quality 3D human and robot motions from text prompts you control it with: → full-body pose keyframes → end-effector positions/rotations → 2D paths and waypoints works on human skeletons and unitree G1 robot plug the outputs directly into mujoco or retarget to other robots using GMR has a web-based interactive demo with a timeline editor. runs locally needs ~17GB VRAM to run inference open source under apache 2.0

NVIDIA DROPPED A MOTION DIFFUSION MODEL FOR HUMANOID ROBOTS trained on 700 hours of mocap data kimodo generates high-quality 3D human and robot motions from text prompts you control it with: → full-body pose keyframes → end-effector positions/rotations → 2D paths and waypoints works on human skeletons and unitree G1 robot plug the outputs directly into mujoco or retarget to other robots using GMR has a web-based interactive demo with a timeline editor. runs locally needs ~17GB VRAM to run inference open source under apache 2.0

Vaishnavi

17,572 görüntüleme • 4 ay önce

The next leap in AI video isn't just better visuals. It's having far more control over how those videos are created. Dreamina Seedance 2.5 is coming soon to the Dreamina platform, and what stands out isn't only the quality it's the workflow built around creators. Here's what's coming: • Multimodal input — combine up to 50 reference assets in a single generation. • Longer video generation — create continuous videos of up to 30 seconds. • Structured control — use white-model and green-screen references for more predictable results. • Multilingual creation — build content for international audiences with localization support. • Targeted refinement — edit specific parts of a video instead of regenerating the entire scene. These upgrades make Dreamina Seedance 2.5 feel less like another prompt-to-video model and more like a complete, controllable AI video production pipeline. Learn more: #dreamina #dreaminapartner #seedance #dreaminaseedance25

The next leap in AI video isn't just better visuals. It's having far more control over how those videos are created. Dreamina Seedance 2.5 is coming soon to the Dreamina platform, and what stands out isn't only the quality it's the workflow built around creators. Here's what's coming: • Multimodal input — combine up to 50 reference assets in a single generation. • Longer video generation — create continuous videos of up to 30 seconds. • Structured control — use white-model and green-screen references for more predictable results. • Multilingual creation — build content for international audiences with localization support. • Targeted refinement — edit specific parts of a video instead of regenerating the entire scene. These upgrades make Dreamina Seedance 2.5 feel less like another prompt-to-video model and more like a complete, controllable AI video production pipeline. Learn more: #dreamina #dreaminapartner #seedance #dreaminaseedance25

Md Riyazuddin

33,164 görüntüleme • 24 gün önce

🧬 We have many foundation models or language models for DNAs, but can we control them? We introduce Ctrl-DNA: Controllable Cell-Type-Specific Regulatory DNA Design via Constrained RL — a reinforcement learning framework for controllable cis-regulatory sequence generation. Paper: Code: 🔬What’s the challenge? Designing regulatory DNA that is both highly expressive in target cell types and inactive in others is essential for synthetic biology, gene therapy, and precision medicine. Yet, controlling these trade-offs is challenging due to sparse, sequence-level rewards and biological constraints. 🔥Why Ctrl-DNA? Ctrl-DNA fine-tunes pre-trained DNA language models using a value model free, Lagrangian-guided RL framework, enabling flexible and customizable constraint optimization. Users can define application-specific thresholds across cell types, balancing expression strength with specificity. ✅ Maximize target-cell expression ✅ Constrain off-target activity under user-defined thresholds ✅ Preserve cell-type-specific TF motif structure Benchmarked on human enhancer and promoter datasets, Ctrl-DNA consistently outperforms prior methods, achieving stronger specificity, higher fitness, and more biologically grounded sequence generation — all with direct control over regulatory trade-offs. Shoutout to the PhD students Xingyu Chen (Xingyu Chen ) and Rex Ma (Rex Ma) for their amazing work leading this project!

🧬 We have many foundation models or language models for DNAs, but can we control them? We introduce Ctrl-DNA: Controllable Cell-Type-Specific Regulatory DNA Design via Constrained RL — a reinforcement learning framework for controllable cis-regulatory sequence generation. Paper: Code: 🔬What’s the challenge? Designing regulatory DNA that is both highly expressive in target cell types and inactive in others is essential for synthetic biology, gene therapy, and precision medicine. Yet, controlling these trade-offs is challenging due to sparse, sequence-level rewards and biological constraints. 🔥Why Ctrl-DNA? Ctrl-DNA fine-tunes pre-trained DNA language models using a value model free, Lagrangian-guided RL framework, enabling flexible and customizable constraint optimization. Users can define application-specific thresholds across cell types, balancing expression strength with specificity. ✅ Maximize target-cell expression ✅ Constrain off-target activity under user-defined thresholds ✅ Preserve cell-type-specific TF motif structure Benchmarked on human enhancer and promoter datasets, Ctrl-DNA consistently outperforms prior methods, achieving stronger specificity, higher fitness, and more biologically grounded sequence generation — all with direct control over regulatory trade-offs. Shoutout to the PhD students Xingyu Chen (Xingyu Chen ) and Rex Ma (Rex Ma) for their amazing work leading this project!

Bo Wang

30,719 görüntüleme • 1 yıl önce