Tom Erez
@nincs.bsky.social
DeepMind, MuJoCo. Love thy neighbor.
Reposted by Tom Erez
Hello all! 👋 🚨 New Preprint Alert! 🚨
Code World Models for General Game-Playing. ♟️🎲 ♣️♥️♠️♦️
I am pleased to announce our new paper, which provides an extremely sample-efficient way to create an agent that can perform well in multi-agent, partially-observed, symbolic environments!
🧵 1/N
Code World Models for General Game-Playing. ♟️🎲 ♣️♥️♠️♦️
I am pleased to announce our new paper, which provides an extremely sample-efficient way to create an agent that can perform well in multi-agent, partially-observed, symbolic environments!
🧵 1/N
October 9, 2025 at 7:27 PM
Hello all! 👋 🚨 New Preprint Alert! 🚨
Code World Models for General Game-Playing. ♟️🎲 ♣️♥️♠️♦️
I am pleased to announce our new paper, which provides an extremely sample-efficient way to create an agent that can perform well in multi-agent, partially-observed, symbolic environments!
🧵 1/N
Code World Models for General Game-Playing. ♟️🎲 ♣️♥️♠️♦️
I am pleased to announce our new paper, which provides an extremely sample-efficient way to create an agent that can perform well in multi-agent, partially-observed, symbolic environments!
🧵 1/N
This is the most interesting contribution to the discussion about consciousness I've read in years. For me, the discussion is pretty much resolved.
Why does the concept of 'red' not feel like the experience of red? Our paper argues this disconnect exists because our deep-seated need to be understood will always exceed the limits of the language we invent to achieve it.
arxiv.org/pdf/2506.12086
#Consciousness #Neuroscience #Philosophy #AI
arxiv.org/pdf/2506.12086
#Consciousness #Neuroscience #Philosophy #AI
arxiv.org
June 17, 2025 at 3:32 PM
This is the most interesting contribution to the discussion about consciousness I've read in years. For me, the discussion is pretty much resolved.
Reposted by Tom Erez
Gemma 3 explained: Longer context, image support, and a new 1B model. → goo.gle/4lV8iaw
Other key enhancements:
🔸 Best model that fits in a single consumer GPU or TPU host
🔸 KV-cache memory reduction with 5-to-1 interleaved attention
🔸 And more!
Read the blog for the full details on Gemma 3.
Other key enhancements:
🔸 Best model that fits in a single consumer GPU or TPU host
🔸 KV-cache memory reduction with 5-to-1 interleaved attention
🔸 And more!
Read the blog for the full details on Gemma 3.
Gemma explained: What’s new in Gemma 3- Google Developers Blog
Google's Gemma 3 model includes vision-language support and architectural changes for resource-friendly multimodal language models.
goo.gle
April 30, 2025 at 9:46 PM
Gemma 3 explained: Longer context, image support, and a new 1B model. → goo.gle/4lV8iaw
Other key enhancements:
🔸 Best model that fits in a single consumer GPU or TPU host
🔸 KV-cache memory reduction with 5-to-1 interleaved attention
🔸 And more!
Read the blog for the full details on Gemma 3.
Other key enhancements:
🔸 Best model that fits in a single consumer GPU or TPU host
🔸 KV-cache memory reduction with 5-to-1 interleaved attention
🔸 And more!
Read the blog for the full details on Gemma 3.
Reposted by Tom Erez
Today we are releasing the dataset of table tennis ball trajectories used to train the Google DeepMind robot that can play amateur table tennis with humans (sites.google.com/corp/view/co...). This work was accepted for presentation at #ICRA2025 and we hope to see you there!
April 30, 2025 at 4:15 PM
Today we are releasing the dataset of table tennis ball trajectories used to train the Google DeepMind robot that can play amateur table tennis with humans (sites.google.com/corp/view/co...). This work was accepted for presentation at #ICRA2025 and we hope to see you there!
Reposted by Tom Erez
Gemma 3 are just amazing models!
but what if you want to manipulate it's internal activations to understand how it does its text generation?
Sascha Rothe is here to teach you how!
Great insights for anyone curious about the inner workings of LLMs!
www.youtube.com/watch?v=JTUs...
but what if you want to manipulate it's internal activations to understand how it does its text generation?
Sascha Rothe is here to teach you how!
Great insights for anyone curious about the inner workings of LLMs!
www.youtube.com/watch?v=JTUs...
Inside Gemma 3: Modifying the output through activation hacking
YouTube video by Google for Developers
www.youtube.com
April 28, 2025 at 1:57 PM
Gemma 3 are just amazing models!
but what if you want to manipulate it's internal activations to understand how it does its text generation?
Sascha Rothe is here to teach you how!
Great insights for anyone curious about the inner workings of LLMs!
www.youtube.com/watch?v=JTUs...
but what if you want to manipulate it's internal activations to understand how it does its text generation?
Sascha Rothe is here to teach you how!
Great insights for anyone curious about the inner workings of LLMs!
www.youtube.com/watch?v=JTUs...
Reposted by Tom Erez
“Wanting to be Understood” - Could a deep human need to be understood be the crucial evolutionary 'gadget' bootstrapping cooperation, culture, and language? We explore this idea using AI simulations in our new paper: arxiv.org/abs/2504.06611 🧠
#Evolution #Cognition #AI #GoogleDeepMind
#Evolution #Cognition #AI #GoogleDeepMind
April 11, 2025 at 11:33 PM
“Wanting to be Understood” - Could a deep human need to be understood be the crucial evolutionary 'gadget' bootstrapping cooperation, culture, and language? We explore this idea using AI simulations in our new paper: arxiv.org/abs/2504.06611 🧠
#Evolution #Cognition #AI #GoogleDeepMind
#Evolution #Cognition #AI #GoogleDeepMind
Reposted by Tom Erez
New paper from our team @GoogleDeepMind!
🚨 We've put LLMs to the test as writing co-pilots – how good are they really at helping us write? LLMs are increasingly used for open-ended tasks like writing assistance, but how do we assess their effectiveness? 🤔
arxiv.org/pdf/2503.19711
🚨 We've put LLMs to the test as writing co-pilots – how good are they really at helping us write? LLMs are increasingly used for open-ended tasks like writing assistance, but how do we assess their effectiveness? 🤔
arxiv.org/pdf/2503.19711
arxiv.org
April 2, 2025 at 9:51 AM
New paper from our team @GoogleDeepMind!
🚨 We've put LLMs to the test as writing co-pilots – how good are they really at helping us write? LLMs are increasingly used for open-ended tasks like writing assistance, but how do we assess their effectiveness? 🤔
arxiv.org/pdf/2503.19711
🚨 We've put LLMs to the test as writing co-pilots – how good are they really at helping us write? LLMs are increasingly used for open-ended tasks like writing assistance, but how do we assess their effectiveness? 🤔
arxiv.org/pdf/2503.19711
Reposted by Tom Erez
🚨 I’m hosting a Student Researcher @GoogleDeepMind!
Join us on the Autonomous Assistants team (led by
@egrefen.bsky.social) to explore multi-agent communication—how agents learn to interact, coordinate, and solve tasks together.
DM me for details!
Join us on the Autonomous Assistants team (led by
@egrefen.bsky.social) to explore multi-agent communication—how agents learn to interact, coordinate, and solve tasks together.
DM me for details!
April 2, 2025 at 9:57 AM
🚨 I’m hosting a Student Researcher @GoogleDeepMind!
Join us on the Autonomous Assistants team (led by
@egrefen.bsky.social) to explore multi-agent communication—how agents learn to interact, coordinate, and solve tasks together.
DM me for details!
Join us on the Autonomous Assistants team (led by
@egrefen.bsky.social) to explore multi-agent communication—how agents learn to interact, coordinate, and solve tasks together.
DM me for details!
Reposted by Tom Erez
🥁Introducing Gemini 2.5, our most intelligent model with impressive capabilities in advanced reasoning and coding.
Now integrating thinking capabilities, 2.5 Pro Experimental is our most performant Gemini model yet. It’s #1 on the LM Arena leaderboard. 🥇
Now integrating thinking capabilities, 2.5 Pro Experimental is our most performant Gemini model yet. It’s #1 on the LM Arena leaderboard. 🥇
March 25, 2025 at 5:25 PM
🥁Introducing Gemini 2.5, our most intelligent model with impressive capabilities in advanced reasoning and coding.
Now integrating thinking capabilities, 2.5 Pro Experimental is our most performant Gemini model yet. It’s #1 on the LM Arena leaderboard. 🥇
Now integrating thinking capabilities, 2.5 Pro Experimental is our most performant Gemini model yet. It’s #1 on the LM Arena leaderboard. 🥇
Reposted by Tom Erez
We all want LLMs to collaborate with humans to help them achieve their goals. But LLMs are not trained to collaborate, they are trained to imitate. Can we teach LM agents to help humans by first making them help each other?
arxiv.org/abs/2503.14481
arxiv.org/abs/2503.14481
Don't lie to your friends: Learning what you know from collaborative self-play
To be helpful assistants, AI agents must be aware of their own capabilities and limitations. This includes knowing when to answer from parametric knowledge versus using tools, when to trust tool outpu...
arxiv.org
March 24, 2025 at 3:39 PM
We all want LLMs to collaborate with humans to help them achieve their goals. But LLMs are not trained to collaborate, they are trained to imitate. Can we teach LM agents to help humans by first making them help each other?
arxiv.org/abs/2503.14481
arxiv.org/abs/2503.14481
Reposted by Tom Erez
Looking for a small or medium sized VLM? PaliGemma 2 spans more than 150x of compute!
Not sure yet if you want to invest the time 🪄finetuning🪄 on your data? Give it a try with our ready-to-use "mix" checkpoints:
🤗 huggingface.co/blog/paligem...
🎤 developers.googleblog.com/en/introduci...
Not sure yet if you want to invest the time 🪄finetuning🪄 on your data? Give it a try with our ready-to-use "mix" checkpoints:
🤗 huggingface.co/blog/paligem...
🎤 developers.googleblog.com/en/introduci...
February 19, 2025 at 5:47 PM
Looking for a small or medium sized VLM? PaliGemma 2 spans more than 150x of compute!
Not sure yet if you want to invest the time 🪄finetuning🪄 on your data? Give it a try with our ready-to-use "mix" checkpoints:
🤗 huggingface.co/blog/paligem...
🎤 developers.googleblog.com/en/introduci...
Not sure yet if you want to invest the time 🪄finetuning🪄 on your data? Give it a try with our ready-to-use "mix" checkpoints:
🤗 huggingface.co/blog/paligem...
🎤 developers.googleblog.com/en/introduci...
Reposted by Tom Erez
How do we ensure humans can still effectively oversee increasingly powerful AI systems? In our blog, we argue that achieving Human-AI complementarity is an underexplored yet vital piece of this puzzle! And, it’s hard, but we achieved it.
🧵(1/10)
🧵(1/10)
December 24, 2024 at 12:01 AM
How do we ensure humans can still effectively oversee increasingly powerful AI systems? In our blog, we argue that achieving Human-AI complementarity is an underexplored yet vital piece of this puzzle! And, it’s hard, but we achieved it.
🧵(1/10)
🧵(1/10)
Reposted by Tom Erez
Making LLMs run efficiently can feel scary, but scaling isn’t magic, it’s math! We wanted to demystify the “systems view” of LLMs and wrote a little textbook called “How To Scale Your Model” which we’re releasing today. 1/n
February 4, 2025 at 6:54 PM
Making LLMs run efficiently can feel scary, but scaling isn’t magic, it’s math! We wanted to demystify the “systems view” of LLMs and wrote a little textbook called “How To Scale Your Model” which we’re releasing today. 1/n
Reposted by Tom Erez
Introducing playground.mujoco.org
Combining MuJoCo’s rich and thriving ecosystem, massively parallel GPU-accelerated simulation, and real-world results across a diverse range of robot platforms: quadrupeds, humanoids, dexterous hands, and arms.
Get started today: pip install playground
Combining MuJoCo’s rich and thriving ecosystem, massively parallel GPU-accelerated simulation, and real-world results across a diverse range of robot platforms: quadrupeds, humanoids, dexterous hands, and arms.
Get started today: pip install playground
MuJoCo Playground
An open-source framework for GPU-accelerated robot learning and sim-to-real transfer
playground.mujoco.org
January 16, 2025 at 8:48 PM
Introducing playground.mujoco.org
Combining MuJoCo’s rich and thriving ecosystem, massively parallel GPU-accelerated simulation, and real-world results across a diverse range of robot platforms: quadrupeds, humanoids, dexterous hands, and arms.
Get started today: pip install playground
Combining MuJoCo’s rich and thriving ecosystem, massively parallel GPU-accelerated simulation, and real-world results across a diverse range of robot platforms: quadrupeds, humanoids, dexterous hands, and arms.
Get started today: pip install playground
Apptronik Partners with Google DeepMind.
apptronik.com/news-collect...
apptronik.com/news-collect...
Apptronik
apptronik.com
December 21, 2024 at 12:33 PM
Apptronik Partners with Google DeepMind.
apptronik.com/news-collect...
apptronik.com/news-collect...
Reposted by Tom Erez
New paper! We show that by using keypoint-based image representation, robot policies become robust to different object types and background changes.
We call this method Prescriptive Point Priors for robot Policies or P3-PO in short. Full project is here: point-priors.github.io
We call this method Prescriptive Point Priors for robot Policies or P3-PO in short. Full project is here: point-priors.github.io
December 10, 2024 at 8:32 PM
New paper! We show that by using keypoint-based image representation, robot policies become robust to different object types and background changes.
We call this method Prescriptive Point Priors for robot Policies or P3-PO in short. Full project is here: point-priors.github.io
We call this method Prescriptive Point Priors for robot Policies or P3-PO in short. Full project is here: point-priors.github.io
Reposted by Tom Erez
HOT 🔥 fastest, most precise, and most capable hand control setup ever...
Less than $450 and fully open-source 🤯
by @huggingface, @therobotstudio, @NepYope
This tendon-driven technology will disrupt robotics! Retweet to accelerate its democratization 🚀
A thread 🧵
Less than $450 and fully open-source 🤯
by @huggingface, @therobotstudio, @NepYope
This tendon-driven technology will disrupt robotics! Retweet to accelerate its democratization 🚀
A thread 🧵
December 15, 2024 at 8:22 AM
HOT 🔥 fastest, most precise, and most capable hand control setup ever...
Less than $450 and fully open-source 🤯
by @huggingface, @therobotstudio, @NepYope
This tendon-driven technology will disrupt robotics! Retweet to accelerate its democratization 🚀
A thread 🧵
Less than $450 and fully open-source 🤯
by @huggingface, @therobotstudio, @NepYope
This tendon-driven technology will disrupt robotics! Retweet to accelerate its democratization 🚀
A thread 🧵
Reposted by Tom Erez
Check out Motivo, a behavioral foundation model for humanoid control by FAIR.
It's a one-of-its-kind unsupervised RL project, and it comes with a demo that is SO fun to play with!
metamotivo.metademolab.com
(for the record, they use compile and cudagraphs -> github.com/facebookrese...)
It's a one-of-its-kind unsupervised RL project, and it comes with a demo that is SO fun to play with!
metamotivo.metademolab.com
(for the record, they use compile and cudagraphs -> github.com/facebookrese...)
December 14, 2024 at 12:44 AM
Check out Motivo, a behavioral foundation model for humanoid control by FAIR.
It's a one-of-its-kind unsupervised RL project, and it comes with a demo that is SO fun to play with!
metamotivo.metademolab.com
(for the record, they use compile and cudagraphs -> github.com/facebookrese...)
It's a one-of-its-kind unsupervised RL project, and it comes with a demo that is SO fun to play with!
metamotivo.metademolab.com
(for the record, they use compile and cudagraphs -> github.com/facebookrese...)
Reposted by Tom Erez
🚀🚀PaliGemma 2 is our updated and improved PaliGemma release using the Gemma 2 models and providing new pre-trained checkpoints for the full cross product of {224px,448px,896px} resolutions and {3B,10B,28B} model sizes.
1/7
1/7
December 5, 2024 at 6:16 PM
🚀🚀PaliGemma 2 is our updated and improved PaliGemma release using the Gemma 2 models and providing new pre-trained checkpoints for the full cross product of {224px,448px,896px} resolutions and {3B,10B,28B} model sizes.
1/7
1/7
Reposted by Tom Erez
The next generation of probabilistic machine learning for weather called GenCast is published in @natureportfolio.bsky.social today 🥳. Amazing to see the collective progress in ML for weather as a field over the last 5 years. 🏖️ www.nature.com/articles/s41...
Probabilistic weather forecasting with machine learning - Nature
GenCast, a probabilistic weather model using artificial intelligence for weather forecasting, has greater skill and speed than the top operational medium-range weather forecast in the world and provid...
www.nature.com
December 4, 2024 at 7:30 PM
The next generation of probabilistic machine learning for weather called GenCast is published in @natureportfolio.bsky.social today 🥳. Amazing to see the collective progress in ML for weather as a field over the last 5 years. 🏖️ www.nature.com/articles/s41...
Reposted by Tom Erez
AI for science could be more impactful than chatbots. It is already helping win Nobel prizes and accelerating drug development and materials discovery.
Today we published an essay about it: why it matters, how it’s happening and its implications. Here is a summary from an econ / social sci lens.
Today we published an essay about it: why it matters, how it’s happening and its implications. Here is a summary from an econ / social sci lens.
November 26, 2024 at 10:39 AM
AI for science could be more impactful than chatbots. It is already helping win Nobel prizes and accelerating drug development and materials discovery.
Today we published an essay about it: why it matters, how it’s happening and its implications. Here is a summary from an econ / social sci lens.
Today we published an essay about it: why it matters, how it’s happening and its implications. Here is a summary from an econ / social sci lens.
If you're coming to Munich for CoRL this week, come say hi!
github.com/google-deepm...
github.com/google-deepm...
MuJoCo meetup at CoRL 2024 · google-deepmind mujoco · Discussion #2203
Hello MuJoCo community! If you are in CoRL 2024 (Munich, Germany), we will be hosting an informal meetup during the 1-2pm lunch break on Friday, November 8th, at the Congress Center's Mars room Swi...
github.com
November 3, 2024 at 6:38 PM
If you're coming to Munich for CoRL this week, come say hi!
github.com/google-deepm...
github.com/google-deepm...
I'm in a starter pack!
go.bsky.app/GZ4hZzu
go.bsky.app/GZ4hZzu
October 28, 2024 at 12:20 PM
I'm in a starter pack!
go.bsky.app/GZ4hZzu
go.bsky.app/GZ4hZzu
Reposted by Tom Erez
I want to take care of everybody and I'm furious at the people who don't want to take care of everybody but I need to not step off the everybody plate just because they suck
September 30, 2024 at 5:41 PM
I want to take care of everybody and I'm furious at the people who don't want to take care of everybody but I need to not step off the everybody plate just because they suck
Reposted by Tom Erez
"A plausible explanation of this is that persons who have achieved or enjoy high social status are less willing to entertain the possibility that a disaster could occur which would spoil everything."
September 11, 2024 at 2:09 PM
"A plausible explanation of this is that persons who have achieved or enjoy high social status are less willing to entertain the possibility that a disaster could occur which would spoil everything."