Charles Sutton
@randomlywalking.bsky.social
Research Scientist, Google DeepMind / Ex-academic / Deep learning to help people write code / ❤️s:🐱🐶☕️🍕
Reposted by Charles Sutton
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
New blog! Advice about career and creativity for researchers and engineers. This time: What my PhD in computer science taught me about gardening. www.theexclusive.org/2025/07/phd-...
What I Learned about Gardening from my PhD in Computer Science
Several years ago, I moved into a house with a garden. Having always lived in apartments, I had never had a garden as an adult. I like having a garden, but I would not say that I like gardening, nor a...
www.theexclusive.org
July 21, 2025 at 9:26 PM
New blog! Advice about career and creativity for researchers and engineers. This time: What my PhD in computer science taught me about gardening. www.theexclusive.org/2025/07/phd-...
Reposted by Charles Sutton
🥁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 Charles Sutton
The inkle round-up:
- Expelled is out and rocking 80+ on metacritic
- A Highland Song is BAFTA nominated…
- … and there’s a sweet disc edition coming from @superraregames.bsky.social next week (27th)
- still some Heaven’s Vault news to come (not a sequel. Not a remaster)
- we’re prototyping
- Expelled is out and rocking 80+ on metacritic
- A Highland Song is BAFTA nominated…
- … and there’s a sweet disc edition coming from @superraregames.bsky.social next week (27th)
- still some Heaven’s Vault news to come (not a sequel. Not a remaster)
- we’re prototyping
SRG#130: A Highland Song (Switch)
Moira McKinnon is running away. A wild adventure through the Scottish Highlands, with open platforming and dynamic storytelling, maps and music in A Highland Song. Each purchase includes the following...
superraregames.com
March 22, 2025 at 10:08 PM
The inkle round-up:
- Expelled is out and rocking 80+ on metacritic
- A Highland Song is BAFTA nominated…
- … and there’s a sweet disc edition coming from @superraregames.bsky.social next week (27th)
- still some Heaven’s Vault news to come (not a sequel. Not a remaster)
- we’re prototyping
- Expelled is out and rocking 80+ on metacritic
- A Highland Song is BAFTA nominated…
- … and there’s a sweet disc edition coming from @superraregames.bsky.social next week (27th)
- still some Heaven’s Vault news to come (not a sequel. Not a remaster)
- we’re prototyping
Reposted by Charles Sutton
I'm happy to advertise an upcoming Student Researcher position on my Agent Frontiers team here at the Google DeepMind Foundational Research Unit, aimed for a start date early in the summer (currently listed as late June, but obviously somewhat flexible).
March 3, 2025 at 2:02 PM
I'm happy to advertise an upcoming Student Researcher position on my Agent Frontiers team here at the Google DeepMind Foundational Research Unit, aimed for a start date early in the summer (currently listed as late June, but obviously somewhat flexible).
Happy fifth birthday to my sourdough starter! Their name is Yeasty.
I learned to bake as a teenager but mostly lapsed after I grew up. When the pandemic hit, I decided to try again, and when there was no yeast, I decided to try sourdough again.
I learned to bake as a teenager but mostly lapsed after I grew up. When the pandemic hit, I decided to try again, and when there was no yeast, I decided to try sourdough again.
March 9, 2025 at 9:05 PM
Happy fifth birthday to my sourdough starter! Their name is Yeasty.
I learned to bake as a teenager but mostly lapsed after I grew up. When the pandemic hit, I decided to try again, and when there was no yeast, I decided to try sourdough again.
I learned to bake as a teenager but mostly lapsed after I grew up. When the pandemic hit, I decided to try again, and when there was no yeast, I decided to try sourdough again.
Reposted by Charles Sutton
As a former Fulbright Scholar (in Chile), I personally understand how terrifying this is. U.S. citizens around the world & foreign nationals in the U.S. — all Fulbright Scholars — have now had their incomes eliminated & are stranded.
DOGE is an illegal, criminal operation that must be shut down.
DOGE is an illegal, criminal operation that must be shut down.
Just got an email from the Fulbright Association. As of right now, funding has been cut off to 12,500 US citizens currently abroad and and more than 7,400 foreigner scholars and students in the United States
March 7, 2025 at 11:34 PM
As a former Fulbright Scholar (in Chile), I personally understand how terrifying this is. U.S. citizens around the world & foreign nationals in the U.S. — all Fulbright Scholars — have now had their incomes eliminated & are stranded.
DOGE is an illegal, criminal operation that must be shut down.
DOGE is an illegal, criminal operation that must be shut down.
Reposted by Charles Sutton
🌟Job ad🌟 We (@gregdnlp.bsky.social, @mattlease.bsky.social and I) are hiring a postdoc fellow within the CosmicAI Institute, to do galactic work with LLMs and generative AI! If you would like to push the frontiers of foundation models to help solve myths of the universe, please apply!
Seeking candidates (within three years of the award of their PhD) for a postdoctoral position with the Explorable Universe research group to perform research on developing next-generation generative AI copilots & agents to aid astronomy research. Info here www.cosmicai.org/jobs/postdoc...
February 25, 2025 at 10:09 PM
🌟Job ad🌟 We (@gregdnlp.bsky.social, @mattlease.bsky.social and I) are hiring a postdoc fellow within the CosmicAI Institute, to do galactic work with LLMs and generative AI! If you would like to push the frontiers of foundation models to help solve myths of the universe, please apply!
This is a magnum opus! Highly recommended.
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 5, 2025 at 5:17 PM
This is a magnum opus! Highly recommended.
Looking back, I think the presents under the tree, “To: Dad, From: Santa”, in Dad’s distinctively illegible handwriting, may have been his way of telling me.
IF YOU BELIEVED in Santa, at what age did you realize he's not real?
December 24, 2024 at 2:03 PM
Looking back, I think the presents under the tree, “To: Dad, From: Santa”, in Dad’s distinctively illegible handwriting, may have been his way of telling me.
Reposted by Charles Sutton
Excited to be able to share what I've been up to recently!
We released a new experimental model, Gemini 2.0 Flash Thinking, which shows the thoughts/reasoning it uses to come up with its answers.
Try it out here! aistudio.google.com/prompts/new_...
We released a new experimental model, Gemini 2.0 Flash Thinking, which shows the thoughts/reasoning it uses to come up with its answers.
Try it out here! aistudio.google.com/prompts/new_...
December 20, 2024 at 1:26 AM
Excited to be able to share what I've been up to recently!
We released a new experimental model, Gemini 2.0 Flash Thinking, which shows the thoughts/reasoning it uses to come up with its answers.
Try it out here! aistudio.google.com/prompts/new_...
We released a new experimental model, Gemini 2.0 Flash Thinking, which shows the thoughts/reasoning it uses to come up with its answers.
Try it out here! aistudio.google.com/prompts/new_...
Reposted by Charles Sutton
We're presenting SWE-agent tomorrow (Wed) at the 11AM poster session, East Exhibit Hall A-C #1000.
We're going to talk about a lot of upcoming SWE-agent features. Join @jyangballin @_carlosejimenez @KLieret and me. I also have a bunch of SWE-agent stickers to hand out :)
We're going to talk about a lot of upcoming SWE-agent features. Join @jyangballin @_carlosejimenez @KLieret and me. I also have a bunch of SWE-agent stickers to hand out :)
December 10, 2024 at 6:16 PM
We're presenting SWE-agent tomorrow (Wed) at the 11AM poster session, East Exhibit Hall A-C #1000.
We're going to talk about a lot of upcoming SWE-agent features. Join @jyangballin @_carlosejimenez @KLieret and me. I also have a bunch of SWE-agent stickers to hand out :)
We're going to talk about a lot of upcoming SWE-agent features. Join @jyangballin @_carlosejimenez @KLieret and me. I also have a bunch of SWE-agent stickers to hand out :)
Reposted by Charles Sutton
The extraordinary recent takeover of ML/AI by #NLP is well-known but insufficiently reflected on.
Look at the @neuripsconf.bsky.social tutorials in 2024!
neurips.cc/virtual/2024...
14 tutorials; 6 have "LLM" in the title; 4 more cover foundation models, with large NLP coverage. That's > 70% 😲
Look at the @neuripsconf.bsky.social tutorials in 2024!
neurips.cc/virtual/2024...
14 tutorials; 6 have "LLM" in the title; 4 more cover foundation models, with large NLP coverage. That's > 70% 😲
NeurIPS 2024 TutorialsNeurIPS 2024
neurips.cc
December 9, 2024 at 7:29 PM
The extraordinary recent takeover of ML/AI by #NLP is well-known but insufficiently reflected on.
Look at the @neuripsconf.bsky.social tutorials in 2024!
neurips.cc/virtual/2024...
14 tutorials; 6 have "LLM" in the title; 4 more cover foundation models, with large NLP coverage. That's > 70% 😲
Look at the @neuripsconf.bsky.social tutorials in 2024!
neurips.cc/virtual/2024...
14 tutorials; 6 have "LLM" in the title; 4 more cover foundation models, with large NLP coverage. That's > 70% 😲
Reposted by Charles Sutton
Do you know what rating you’ll give after reading the intro? Are your confidence scores 4 or higher? Do you not respond in rebuttal phases? Are you worried how it will look if your rating is the only 8 among 3’s? This thread is for you.
November 27, 2024 at 5:25 PM
Do you know what rating you’ll give after reading the intro? Are your confidence scores 4 or higher? Do you not respond in rebuttal phases? Are you worried how it will look if your rating is the only 8 among 3’s? This thread is for you.
Benchmarks drive the decision making for large modeling efforts. Here are some great thoughts about how to design good benchmarks!
I wrote some thoughts on how to build good LM benchmarks: ofir.io/How-to-Build...
How to Build Good Language Modeling Benchmarks
Building benchmarks is important because they shine a spotlight on the weaknesses of existing language models and so can guide the community on how to improve them.
ofir.io
November 26, 2024 at 2:51 AM
Benchmarks drive the decision making for large modeling efforts. Here are some great thoughts about how to design good benchmarks!
That’s right. You might think that all successful CS academics are good at running. But that’s only because the ones who weren’t, have been eaten by bears.
A disproportionate number of sucessful CS academics have some intense cardio hobby. Took me some years to understand.
Every time I see someone post this image it goes viral
November 21, 2024 at 12:45 AM
That’s right. You might think that all successful CS academics are good at running. But that’s only because the ones who weren’t, have been eaten by bears.
Reposted by Charles Sutton
Starter pack for University of Edinburgh researchers done by the amazing ramandutt4.bsky.social - go.bsky.app/KRNDkN7
University of Edinburgh Starter Pack
Join the conversation
go.bsky.app
November 20, 2024 at 4:39 PM
Starter pack for University of Edinburgh researchers done by the amazing ramandutt4.bsky.social - go.bsky.app/KRNDkN7
Reposted by Charles Sutton
I did an unscientific, uncontrolled experiment for #EMNLP2024—details in 🧵👇. I posted my conference & workshop papers to 5 socials. Clear results: Mastodon is near dead, Threads may have users but not my people, not giving up on X/Twitter yet, but Bluesky is worth investing in.
November 18, 2024 at 6:40 PM
I did an unscientific, uncontrolled experiment for #EMNLP2024—details in 🧵👇. I posted my conference & workshop papers to 5 socials. Clear results: Mastodon is near dead, Threads may have users but not my people, not giving up on X/Twitter yet, but Bluesky is worth investing in.
Reposted by Charles Sutton
Our team is looking for an ML hacker extraordinaire. make the TPUs hum with the sound of AGI. Job description: make intelligence = f(compute) unbounded : boards.greenhouse.io/deepmind/job...
Research Engineer, Generative AI
London, UK
boards.greenhouse.io
November 16, 2024 at 4:00 PM
Our team is looking for an ML hacker extraordinaire. make the TPUs hum with the sound of AGI. Job description: make intelligence = f(compute) unbounded : boards.greenhouse.io/deepmind/job...
Reposted by Charles Sutton
Wanted to share that Varun Godbole recently released a prompting playbook. The title says prompt tuning, but this is text prompts, not soft prompts.
github.com/varungodbole...
github.com/varungodbole...
GitHub - varungodbole/prompt-tuning-playbook: A playbook for effectively prompting post-trained LLMs
A playbook for effectively prompting post-trained LLMs - varungodbole/prompt-tuning-playbook
github.com
November 11, 2024 at 3:51 PM
Wanted to share that Varun Godbole recently released a prompting playbook. The title says prompt tuning, but this is text prompts, not soft prompts.
github.com/varungodbole...
github.com/varungodbole...
These are great starter packs!
If you like these, you should check out this starter pack of DeepMind researchers: go.bsky.app/GZ4hZzu
If you like these, you should check out this starter pack of DeepMind researchers: go.bsky.app/GZ4hZzu
New here? Interested in AI/ML? Check out these great starter packs!
AI: go.bsky.app/SipA7it
RL: go.bsky.app/3WPHcHg
Women in AI: go.bsky.app/LaGDpqg
NLP: go.bsky.app/SngwGeS
AI and news: go.bsky.app/5sFqVNS
You can also search all starter packs here: blueskydirectory.com/starter-pack...
AI: go.bsky.app/SipA7it
RL: go.bsky.app/3WPHcHg
Women in AI: go.bsky.app/LaGDpqg
NLP: go.bsky.app/SngwGeS
AI and news: go.bsky.app/5sFqVNS
You can also search all starter packs here: blueskydirectory.com/starter-pack...
November 10, 2024 at 9:13 PM
These are great starter packs!
If you like these, you should check out this starter pack of DeepMind researchers: go.bsky.app/GZ4hZzu
If you like these, you should check out this starter pack of DeepMind researchers: go.bsky.app/GZ4hZzu
Reposted by Charles Sutton
Hi! Are you new here and looking for more cat accounts to follow? Here is a starter pack I made of cat accounts that use ALT text and are run by real people - no bots or image scrapers! go.bsky.app/J8yqxnN
November 8, 2024 at 1:35 AM
Hi! Are you new here and looking for more cat accounts to follow? Here is a starter pack I made of cat accounts that use ALT text and are run by real people - no bots or image scrapers! go.bsky.app/J8yqxnN
Our team has been working hard to harness the power of AI to make software more secure.✨🔐
Today we are excited to share a major milestone: our AI agent has discovered its first real-world security vulnerability!
Read here for more:
googleprojectzero.blogspot.com/2024/10/from...
Today we are excited to share a major milestone: our AI agent has discovered its first real-world security vulnerability!
Read here for more:
googleprojectzero.blogspot.com/2024/10/from...
From Naptime to Big Sleep: Using Large Language Models To Catch Vulnerabilities In Real-World Code
Posted by the Big Sleep team Introduction In our previous post, Project Naptime: Evaluating Offensive Security Capabilities of Large L...
googleprojectzero.blogspot.com
November 1, 2024 at 8:32 PM
Our team has been working hard to harness the power of AI to make software more secure.✨🔐
Today we are excited to share a major milestone: our AI agent has discovered its first real-world security vulnerability!
Read here for more:
googleprojectzero.blogspot.com/2024/10/from...
Today we are excited to share a major milestone: our AI agent has discovered its first real-world security vulnerability!
Read here for more:
googleprojectzero.blogspot.com/2024/10/from...
A black cat crossed my path on Halloween! I think this means I get twice as much good luck!
November 1, 2024 at 5:13 AM
A black cat crossed my path on Halloween! I think this means I get twice as much good luck!
I used to have a joke about Heraclitus, but I can’t tell it twice
I have a joke about Procrustes, but I’d have to cut it short
I have a joke about Helen, but that ship has sailed.
October 24, 2024 at 3:57 AM
I used to have a joke about Heraclitus, but I can’t tell it twice