正在加载视频...

视频加载失败

返回首页

Every wondered if we can model motion as a language? can we tokenize this new language? is it useful? Turns out tremendously! 🚀 In out latest #NeurIPS2024 paper on QueST: Self-Supervised Skill Abstractions for Learning Continuous Control, we find that action tokenization matters a lot! We can learn skill...

animesh_garg's profile picture

Animesh Garg

33,626 subscribers

26,218 次观看 • 1 年前 •via X (Twitter)

科学技术教育#NeurIPS2024
LIVE

Anya Rossi• Live Now

Private livecam show

10 条评论

Animesh Garg 的头像
Animesh Garg1 年前

also a shout out to @LerrelPinto and team for their work on BAKU which is concurrent and equally impressive in terms of thinning the chaff in BC architectures. This is important and their ideas are also very timely. It is exciting to see the field simplify architectures and also simpler latent variable transformer architectures becoming very competitive at multimodal action space modeling.

Animesh Garg 的头像
Animesh Garg1 年前

Also shout out to @zsoltkira @AlexToshev and team for another learned action tokenizer “…show that a learned tokenization allows for sufficient modeling precision, yielding the best performance on downstream tasks. For discrete actions, we demonstrate that semantically aligning these actions with the native output token space of the MLLM leads to the strongest performance. We arrive at these lessons via a thorough study of seven action space adapters…” Learned action tokenization, combined with simplified transformer architectures is scaling very well I’m so glad to see more activity on this question and that too from my next-door neighbor @zsoltkira

Dr. Angelica Lim @petitegeek.bsky.social 的头像
Dr. Angelica Lim @petitegeek.bsky.social1 年前

100% agree! Tokenizing motion as a language also helps for translating language to motion (IROS 2022) We also have relevant work in BC with action tokens coming up at IROS 2024

Animesh Garg 的头像
Animesh Garg1 年前

thanks for sharing. we will look at it and definitely cite it in the camera ready.

Kaustubh Sridhar 的头像
Kaustubh Sridhar1 年前

Hi @animesh_garg @MeteAtharva , this is a really interesting paper! Just wanted to share our ICLR 24 paper, MCNN which also previously improved over diffusion policy and other BC baselines. Thanks

Ge Yang 的头像
Ge Yang1 年前

This is really good!

You Jiacheng 的头像
You Jiacheng1 年前

I feel the effectiveness attribution of the causal tokenizer design might be wrong. Since the AR transformer generates the tokens auto-regressively (and causally), it's natural to use a causal tokenizer -- otherwise there is a mismatch.

wildiris 的头像
wildiris1 年前

This is fantastic. It's exactly what I'm working on in hardware.

Animesh Garg 的头像
Animesh Garg1 年前

Great to hear that would love to know more how to do that in hardware

Anonymous Deep Learning 的头像
Anonymous Deep Learning1 年前

Cool but is it really necessary to reduce motion information to a much lower bandwidth information such as language? For humans language is easy to understand but robots can understand trajectories in coordinates. So isn’t language a shortcut here?

相关视频

Goal-conditioned RL (GCRL) is great - unsupervised, can use data (in offline mode), flexibility to define tasks at test time. But can we run GCRL on *language data*?? In our new work we show that language GCRL enables sophisticated test-time reasoning for interactive tasks! 🧵👇
0:22

Goal-conditioned RL (GCRL) is great - unsupervised, can use data (in offline mode), flexibility to define tasks at test time. But can we run GCRL on *language data*?? In our new work we show that language GCRL enables sophisticated test-time reasoning for interactive tasks! 🧵👇

Sergey Levine

18,782 次观看 • 1 年前

MotionGPT: Human Motion as a Foreign Language paper page: Though the advancement of pre-trained large language models unfolds, the exploration of building a unified model for language and other multi-modal data, such as motion, remains challenging and untouched so far. Fortunately, human motion displays a semantic coupling akin to human language, often perceived as a form of body language. By fusing language data with large-scale motion models, motion-language pre-training that can enhance the performance of motion-related tasks becomes feasible. Driven by this insight, we propose MotionGPT, a unified, versatile, and user-friendly motion-language model to handle multiple motion-relevant tasks. Specifically, we employ the discrete vector quantization for human motion and transfer 3D motion into motion tokens, similar to the generation process of word tokens. Building upon this "motion vocabulary", we perform language modeling on both motion and text in a unified manner, treating human motion as a specific language. Moreover, inspired by prompt learning, we pre-train MotionGPT with a mixture of motion-language data and fine-tune it on prompt-based question-and-answer tasks. Extensive experiments demonstrate that MotionGPT achieves state-of-the-art performances on multiple motion tasks including text-driven motion generation, motion captioning, motion prediction, and motion in-between.
0:48

MotionGPT: Human Motion as a Foreign Language paper page: Though the advancement of pre-trained large language models unfolds, the exploration of building a unified model for language and other multi-modal data, such as motion, remains challenging and untouched so far. Fortunately, human motion displays a semantic coupling akin to human language, often perceived as a form of body language. By fusing language data with large-scale motion models, motion-language pre-training that can enhance the performance of motion-related tasks becomes feasible. Driven by this insight, we propose MotionGPT, a unified, versatile, and user-friendly motion-language model to handle multiple motion-relevant tasks. Specifically, we employ the discrete vector quantization for human motion and transfer 3D motion into motion tokens, similar to the generation process of word tokens. Building upon this "motion vocabulary", we perform language modeling on both motion and text in a unified manner, treating human motion as a specific language. Moreover, inspired by prompt learning, we pre-train MotionGPT with a mixture of motion-language data and fine-tune it on prompt-based question-and-answer tasks. Extensive experiments demonstrate that MotionGPT achieves state-of-the-art performances on multiple motion tasks including text-driven motion generation, motion captioning, motion prediction, and motion in-between.

AK

125,319 次观看 • 3 年前

Self-supervised representation learning looks a bit like RL. What if we literally use RL as a SSL method for visual representations? Turns out that it works quite well. In new work by Dibya Ghosh, we show how this can be done:
1:09

Self-supervised representation learning looks a bit like RL. What if we literally use RL as a SSL method for visual representations? Turns out that it works quite well. In new work by Dibya Ghosh, we show how this can be done:

Sergey Levine

48,715 次观看 • 1 年前

Today, we announced 𝗥𝗧-𝟮: a first of its kind vision-language-action model to control robots. 🤖 It learns from both web and robotics data and translates this knowledge into generalised instructions. Find out more:
0:26

Today, we announced 𝗥𝗧-𝟮: a first of its kind vision-language-action model to control robots. 🤖 It learns from both web and robotics data and translates this knowledge into generalised instructions. Find out more:

Google DeepMind

537,808 次观看 • 3 年前

📢SUPER excited to present NeO 360, our #ICCV23 work on sparse view 360-degree view synthesis of outdoor scenes! Presenting a large-scale dataset and a strong baseline for this challenging task 🧵👇 Webpage: Robotics@GT Georgia Tech School of Interactive Computing Machine Learning at Georgia Tech Toyota Research Institute (TRI)
0:28

📢SUPER excited to present NeO 360, our #ICCV23 work on sparse view 360-degree view synthesis of outdoor scenes! Presenting a large-scale dataset and a strong baseline for this challenging task 🧵👇 Webpage: Robotics@GT Georgia Tech School of Interactive Computing Machine Learning at Georgia Tech Toyota Research Institute (TRI)

Zubair Irshad

51,624 次观看 • 2 年前

Milestone! We (robotic arms for gadgets assembly) finished the first commercial order, which brought the first revenue. Here are some learnings from this: The customer was a smart toy manufacturer. The task was to add a heatsink to Raspberry Pi. We received parts from them and returned the assembled modules back. Currently, it's done by teleoperation. Later it will be done by a remote employee via the Internet. Then it will be automated action by action, reducing the operator's time on this and making the task profitable. ps. If you have an assembly task that we can do for you asynchronically - leave a comment below. Learning 1. It's possible! This task which is usually done by the human arm with 5 fingers can be done with a two-finger gripper with the addition of a couple of simple tooling. The task was not simplified. We peeled off thin films from stickers, unpacked paper boxes, moved PCB boards full of components, etc. And no unsolvable problems have been encountered yet. Challenges: 1) The paper box shifted during the opening Solved with the plastic walls that you can lean against 2) Heat pad, stuck to the gripper instead of heat sync. Can be solved by gripper with a pump, but this time solved with the patience of the operator 3) The film on the pad is very thin. Turned out that sub-millimeter arm precision is enough to peel it off with just a regular gripper. 4) The working area has not enough space. You'll only know this by doing real tasks in bulk. This could be solved by an extra pair of long arms, but in this case, solved with the patience of the operator. I think that in the end, we will have 5-10 types of universal tooling and 5-10 types of grippers to solve almost all the problems in such assembly tasks. Learning 2. It's slow. It took 5 times more time, than doing it with human hands. But the good news is there's a lot of room for improvement. We now have specific “time for task” metrics, which we will decrease with iterations. The main reasons for slowness: 1) To rotate the gripper to a steep angle you are forced to control one robot arm with two hands instead of using both arms. We can fix this by just making more room for rotations. 2) Grabbing PCB board with two arms is hard. A slight difference in rotation can break the board, and it's hard to control these angles visually. To solve this, the best way is to use force feedback so you can feel the pressure applied to the item. 3) Accuracy and steadiness is still can be improved We will try a metal version and double the motors to do this. 4) It is physically difficult for the human hands to move with such precision To solve this, we will add a pad for the hands like in surgical robots Learning 3. It's a good business model The "Factory in the cloud" is a good business model for this stage. You send us parts and we send back assembled modules. Currently, it's more convenient than sending a robot to your place, as we can iterate/fix the robot quickly and utilize it 100% of the time. When we polish the set-up over time - we can send robots to your place. So if we can assemble something for you in the USA with Chinese prices by using modern automation - leave a comment below.
0:48

Milestone! We (robotic arms for gadgets assembly) finished the first commercial order, which brought the first revenue. Here are some learnings from this: The customer was a smart toy manufacturer. The task was to add a heatsink to Raspberry Pi. We received parts from them and returned the assembled modules back. Currently, it's done by teleoperation. Later it will be done by a remote employee via the Internet. Then it will be automated action by action, reducing the operator's time on this and making the task profitable. ps. If you have an assembly task that we can do for you asynchronically - leave a comment below. Learning 1. It's possible! This task which is usually done by the human arm with 5 fingers can be done with a two-finger gripper with the addition of a couple of simple tooling. The task was not simplified. We peeled off thin films from stickers, unpacked paper boxes, moved PCB boards full of components, etc. And no unsolvable problems have been encountered yet. Challenges: 1) The paper box shifted during the opening Solved with the plastic walls that you can lean against 2) Heat pad, stuck to the gripper instead of heat sync. Can be solved by gripper with a pump, but this time solved with the patience of the operator 3) The film on the pad is very thin. Turned out that sub-millimeter arm precision is enough to peel it off with just a regular gripper. 4) The working area has not enough space. You'll only know this by doing real tasks in bulk. This could be solved by an extra pair of long arms, but in this case, solved with the patience of the operator. I think that in the end, we will have 5-10 types of universal tooling and 5-10 types of grippers to solve almost all the problems in such assembly tasks. Learning 2. It's slow. It took 5 times more time, than doing it with human hands. But the good news is there's a lot of room for improvement. We now have specific “time for task” metrics, which we will decrease with iterations. The main reasons for slowness: 1) To rotate the gripper to a steep angle you are forced to control one robot arm with two hands instead of using both arms. We can fix this by just making more room for rotations. 2) Grabbing PCB board with two arms is hard. A slight difference in rotation can break the board, and it's hard to control these angles visually. To solve this, the best way is to use force feedback so you can feel the pressure applied to the item. 3) Accuracy and steadiness is still can be improved We will try a metal version and double the motors to do this. 4) It is physically difficult for the human hands to move with such precision To solve this, we will add a pad for the hands like in surgical robots Learning 3. It's a good business model The "Factory in the cloud" is a good business model for this stage. You send us parts and we send back assembled modules. Currently, it's more convenient than sending a robot to your place, as we can iterate/fix the robot quickly and utilize it 100% of the time. When we polish the set-up over time - we can send robots to your place. So if we can assemble something for you in the USA with Chinese prices by using modern automation - leave a comment below.

Igor Kulakov

37,266 次观看 • 1 年前

We are back. After one year of quiet building. Introducing GENE-26.5, our first robotic brain that takes a major step toward human-level capability. For years, robotics has struggled to learn from the world’s largest and valuable data source: Humans. Solving it means rethinking the whole stack from the ground up: - A robotics-native foundation model. - A 1:1 human-like robotic hand. - A noninvasive data collection glove for motion, force, and touch. - A simulator that turns weeks of experiments into minutes. GENE-26.5 is trained across language, vision, proprioception, tactile, and action. We designed a set of tasks to test how far we can go with this new paradigm. Fully autonomous, 1x speed, one model, same weights. (Enjoy with sound on) We are approaching the endgame for robotics. And this is just a beginning.
3:05

We are back. After one year of quiet building. Introducing GENE-26.5, our first robotic brain that takes a major step toward human-level capability. For years, robotics has struggled to learn from the world’s largest and valuable data source: Humans. Solving it means rethinking the whole stack from the ground up: - A robotics-native foundation model. - A 1:1 human-like robotic hand. - A noninvasive data collection glove for motion, force, and touch. - A simulator that turns weeks of experiments into minutes. GENE-26.5 is trained across language, vision, proprioception, tactile, and action. We designed a set of tasks to test how far we can go with this new paradigm. Fully autonomous, 1x speed, one model, same weights. (Enjoy with sound on) We are approaching the endgame for robotics. And this is just a beginning.

Genesis AI

2,711,060 次观看 • 2 个月前

Robot AI brains, aka Vision-Language-Action models, cannot adapt to new tasks as easily as LLMs like Gemini, ChatGPT, or Grok. LLMs can adapt quickly with their in-context learning (ICL) capabilities. But can we inject ICL abilities into a pre-trained VLA like pi0? Yes! Introducing RICL (Retraining for In-Context Learning), our Conference on Robot Learning (CoRL) 2025 paper. Our RICL-pi0 model can adapt to unseen objects, novel motions, and new scenes with just ICL and RAG (retrieval-augmented generation). RICL-pi0 also boosts performance on the long-tail of tasks. A quick 1 minute video summary:
1:03

Robot AI brains, aka Vision-Language-Action models, cannot adapt to new tasks as easily as LLMs like Gemini, ChatGPT, or Grok. LLMs can adapt quickly with their in-context learning (ICL) capabilities. But can we inject ICL abilities into a pre-trained VLA like pi0? Yes! Introducing RICL (Retraining for In-Context Learning), our Conference on Robot Learning (CoRL) 2025 paper. Our RICL-pi0 model can adapt to unseen objects, novel motions, and new scenes with just ICL and RAG (retrieval-augmented generation). RICL-pi0 also boosts performance on the long-tail of tasks. A quick 1 minute video summary:

Kaustubh Sridhar

52,158 次观看 • 10 个月前

Our newest model, π0.7, has some interesting emergent capabilities: it can control a new robot to fold shirts for which we had no shirt folding data, figure out how to use an appliance with language-based coaching, and perform a wide range of dexterous tasks all in one model!
4:02

Our newest model, π0.7, has some interesting emergent capabilities: it can control a new robot to fold shirts for which we had no shirt folding data, figure out how to use an appliance with language-based coaching, and perform a wide range of dexterous tasks all in one model!

Physical Intelligence

459,547 次观看 • 2 个月前

We can teach LLMs to write better robot code through natural language feedback. But can LLMs remember what they were taught and improve their teachability over time? Introducing our latest work, Learning to Learn Faster from Human Feedback with Language Model Predictive Control
0:40

We can teach LLMs to write better robot code through natural language feedback. But can LLMs remember what they were taught and improve their teachability over time? Introducing our latest work, Learning to Learn Faster from Human Feedback with Language Model Predictive Control

Jacky Liang

86,652 次观看 • 2 年前

The holy grail of robotics is to be able to perform previously-unseen, out-of-distribution manipulation tasks “zero shot” in a new environment. NovaFlow proposes an approach which (1) generates a video, (2) computes predicted flow — how points move through the scene — and (3) uses this flow as an objective to generate a motion. Using this procedure, NovaFlow generates motions in unseen scenes, for unseen tasks, and can transfer across embodiments. To learn more, we are joined by Hongyu Li and Jiahui Fu from RAI. Watch Episode #63 of RoboPapers with Chris Paxton and Michael Cho - Rbt/Acc now to learn more!
1:08:49

The holy grail of robotics is to be able to perform previously-unseen, out-of-distribution manipulation tasks “zero shot” in a new environment. NovaFlow proposes an approach which (1) generates a video, (2) computes predicted flow — how points move through the scene — and (3) uses this flow as an objective to generate a motion. Using this procedure, NovaFlow generates motions in unseen scenes, for unseen tasks, and can transfer across embodiments. To learn more, we are joined by Hongyu Li and Jiahui Fu from RAI. Watch Episode #63 of RoboPapers with Chris Paxton and Michael Cho - Rbt/Acc now to learn more!

RoboPapers

21,295 次观看 • 4 个月前

Everyone knows action chunking is great for imitation learning. It turns out that we can extend its success to RL to better leverage prior data for improved exploration and online sample efficiency! The recipe to achieve this is incredibly simple. 🧵 1/N
0:25

Everyone knows action chunking is great for imitation learning. It turns out that we can extend its success to RL to better leverage prior data for improved exploration and online sample efficiency! The recipe to achieve this is incredibly simple. 🧵 1/N

Qiyang (Colin) Li

48,231 次观看 • 1 年前

We are excited to announce the 1st version of our multimodal assistant, Yasa-1, a language assistant with visual and auditory sensors that can take actions via code execution 🪄. Yasa-1 can understand text, images, videos, sounds & more! 🚀 Check out more details below👇
0:45

We are excited to announce the 1st version of our multimodal assistant, Yasa-1, a language assistant with visual and auditory sensors that can take actions via code execution 🪄. Yasa-1 can understand text, images, videos, sounds & more! 🚀 Check out more details below👇

Reka

814,187 次观看 • 2 年前

Introducing the USC Physical Superintelligence (PSI) Lab ( We are rebranding to better reflect our current focus. From here on out, we are tackling one thing: solving robotics and physical intelligence with every model, every bug, and every line of code. And yes, we are hiring at all levels, especially PhDs in this cycle and potential PostDocs who are excited about robotics. We hope you can join us in this journey! 1/9
4:14

Introducing the USC Physical Superintelligence (PSI) Lab ( We are rebranding to better reflect our current focus. From here on out, we are tackling one thing: solving robotics and physical intelligence with every model, every bug, and every line of code. And yes, we are hiring at all levels, especially PhDs in this cycle and potential PostDocs who are excited about robotics. We hope you can join us in this journey! 1/9

Yue Wang

25,456 次观看 • 7 个月前

The robot is learning several novel tasks instantly, after just ONE demonstration each... Instant Policy makes it possible: no extra training, no weight updates, just pure in-context learning. It just got accepted at ICLR 2025, and it’s changing how robots learn. With just a single demo, a robot can pick up a new task and start performing it right away. Why this is a big deal: ✅ Learns tasks instantly with just one or a few demonstrations ✅ Improves over time as more demonstrations are given ✅ Uses simulation-based training with “pseudo-demonstrations” for scalability ✅ Can transfer skills across different robots and even follow language-defined tasks It brings in-context learning to robotics, opening up new possibilities for flexible, real-world automation. You can try it yourself: code and weights are available at • • • • Thank you to Edward Johns, Director of the Robot Learning Lab at Imperial College for sharing their work! 🙏
0:50

The robot is learning several novel tasks instantly, after just ONE demonstration each... Instant Policy makes it possible: no extra training, no weight updates, just pure in-context learning. It just got accepted at ICLR 2025, and it’s changing how robots learn. With just a single demo, a robot can pick up a new task and start performing it right away. Why this is a big deal: ✅ Learns tasks instantly with just one or a few demonstrations ✅ Improves over time as more demonstrations are given ✅ Uses simulation-based training with “pseudo-demonstrations” for scalability ✅ Can transfer skills across different robots and even follow language-defined tasks It brings in-context learning to robotics, opening up new possibilities for flexible, real-world automation. You can try it yourself: code and weights are available at • • • • Thank you to Edward Johns, Director of the Robot Learning Lab at Imperial College for sharing their work! 🙏

Ilir Aliu

46,285 次观看 • 1 年前

Honorable please check out our work. We are ready to put this in parliament, we just need honest partners. Attached is an example of converting Twi audio into English - so you can speak your language and everyone else still can follow More examples on my profile
1:59

Honorable please check out our work. We are ready to put this in parliament, we just need honest partners. Attached is an example of converting Twi audio into English - so you can speak your language and everyone else still can follow More examples on my profile

Paul Azunre

98,848 次观看 • 1 年前

Chain-of-thought reasoning is a powerful tool to enable language models to work through complex problems. Can we use this with robots? With embodied chain-of-thought, vision-language-action (VLA) models can think through perception and planning! A 🧵👇
1:39

Chain-of-thought reasoning is a powerful tool to enable language models to work through complex problems. Can we use this with robots? With embodied chain-of-thought, vision-language-action (VLA) models can think through perception and planning! A 🧵👇

Sergey Levine

30,388 次观看 • 2 年前

Excited to finally share Generative Value Learning (GVL), my Google DeepMind project on extracting universal value functions from long-context VLMs via in-context learning! We discovered a simple method to generate zero-shot and few-shot values for 300+ robot tasks and 50+ datasets using SOTA VLMs like Gemini (Try out the demo on our website on your robot video today!) I worked a lot on leveraging foundation models as guidance for robots in my PhD, and to me, this result forges a new frontier in how we can use foundation models for robot learning, given its broad applicability independent of embodiment and task types. Quite excited about how we can build on this work as a community!
0:32

Excited to finally share Generative Value Learning (GVL), my Google DeepMind project on extracting universal value functions from long-context VLMs via in-context learning! We discovered a simple method to generate zero-shot and few-shot values for 300+ robot tasks and 50+ datasets using SOTA VLMs like Gemini (Try out the demo on our website on your robot video today!) I worked a lot on leveraging foundation models as guidance for robots in my PhD, and to me, this result forges a new frontier in how we can use foundation models for robot learning, given its broad applicability independent of embodiment and task types. Quite excited about how we can build on this work as a community!

Jason Ma

98,090 次观看 • 1 年前

DOBOT robotics' humanoid robot is making a big leap in industrial applications, achieving cross-scenario, multi-task generalization. This is thanks to two major breakthroughs: efficient human-to-robot motion mapping and knowledge-driven generative VLA tech. ► The robot handles complex tasks like precision assembly and can work reliably in temperatures over 50°C (122°F). ► It excels at multi-robot collaborative tasks and can adaptively grasp soft or irregularly shaped objects. With a repeatability of ±0.05mm, this model demonstrates strong adaptability and reliability in dynamic industrial settings. It's now being applied in warehouse anomaly handling and quality inspection.
2:09

DOBOT robotics' humanoid robot is making a big leap in industrial applications, achieving cross-scenario, multi-task generalization. This is thanks to two major breakthroughs: efficient human-to-robot motion mapping and knowledge-driven generative VLA tech. ► The robot handles complex tasks like precision assembly and can work reliably in temperatures over 50°C (122°F). ► It excels at multi-robot collaborative tasks and can adaptively grasp soft or irregularly shaped objects. With a repeatability of ±0.05mm, this model demonstrates strong adaptability and reliability in dynamic industrial settings. It's now being applied in warehouse anomaly handling and quality inspection.

RoboHub🤖

91,953 次观看 • 10 个月前

We are excited to announce CLIN 🤖: The first continually learning language agent that excels in both task adaptation and generalization to unseen tasks and environments in a pure zero-shot setup. Aristo Team at Ai2 Ai2 Website: Let's dive in 🧵 (1/n)
0:49

We are excited to announce CLIN 🤖: The first continually learning language agent that excels in both task adaptation and generalization to unseen tasks and environments in a pure zero-shot setup. Aristo Team at Ai2 Ai2 Website: Let's dive in 🧵 (1/n)

Bodhisattwa Majumder

35,052 次观看 • 2 年前