Загрузка видео...

Не удалось загрузить видео

На главную

Today we’re introducing MIRA, a new multiplayer world model, built with General Intuition, in collaboration with Epic Games. We release an in-depth technical report, dataset, as well as an online demo that you can try right now (link below).

kyutai_labs's profile picture

kyutai

27,698 subscribers

75,757 просмотров • 1 месяц назад •via X (Twitter)

Наука и технологииОбразование
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Introducing ABot-World, the latest world model running on Reactor by AMAP from Alibaba Group. Explore infinite worlds, with new scenes and dynamics generating in real time as you move. Try the demo now:
0:15

Introducing ABot-World, the latest world model running on Reactor by AMAP from Alibaba Group. Explore infinite worlds, with new scenes and dynamics generating in real time as you move. Try the demo now:

reactor

23,079 просмотров • 1 месяц назад

We've got some major news that we couldn't wait to *share* with you... As of today, you can now curate and share your notebooks with ANYONE using a public link. Do you have an amazing notebook you want to share with the community? Drop the link below!
0:16

We've got some major news that we couldn't wait to *share* with you... As of today, you can now curate and share your notebooks with ANYONE using a public link. Do you have an amazing notebook you want to share with the community? Drop the link below!

Gemini Notebook

244,213 просмотров • 1 год назад

We released a new model 🎉 !!! Firefly Image 2 !!!! As a scientist I am in disbelief that such model is possible with a small dataset as we had. We compensated artists for training it. We worked very hard on it. Please, enjoy! #AdobeMax
0:16

We released a new model 🎉 !!! Firefly Image 2 !!!! As a scientist I am in disbelief that such model is possible with a small dataset as we had. We compensated artists for training it. We worked very hard on it. Please, enjoy! #AdobeMax

Kris Kashtanova

200,492 просмотров • 2 лет назад

New gameplay for Project Orc, an open-world survival action RPG where you play as an orc, building and leading clans in a harsh fantasy world. It features multiplayer modes with dedicated servers for co-op or MMO-style play.
0:15

New gameplay for Project Orc, an open-world survival action RPG where you play as an orc, building and leading clans in a harsh fantasy world. It features multiplayer modes with dedicated servers for co-op or MMO-style play.

Pirat_Nation 🔴

307,606 просмотров • 6 месяцев назад

describing an aesthetic in a prompt can be tough, so we made a button for it introducing Design Variations instantly generate, explore, and apply beautiful new UI layouts with a single click try it today in AI Studio
0:16

describing an aesthetic in a prompt can be tough, so we made a button for it introducing Design Variations instantly generate, explore, and apply beautiful new UI layouts with a single click try it today in AI Studio

Google AI Studio

237,925 просмотров • 2 месяцев назад

So this is the dream: A video world model that takes an image as input and renders an environment you can explore and interact with. It could be a constant video stream - like your own lofi girl! Or you could jump in and “play” as a character.
0:15

So this is the dream: A video world model that takes an image as input and renders an environment you can explore and interact with. It could be a constant video stream - like your own lofi girl! Or you could jump in and “play” as a character.

Justine Moore

95,492 просмотров • 1 год назад

Opening doors to Headout's new HQ in BLR. A space we imagined as a Museum of the World, designed as an ode to the big, beautiful world around us. We made it with love, and there’s room for many more. We’re hiring across roles. Come build with us.
0:10

Opening doors to Headout's new HQ in BLR. A space we imagined as a Museum of the World, designed as an ode to the big, beautiful world around us. We made it with love, and there’s room for many more. We’re hiring across roles. Come build with us.

Varun Khona

43,094 просмотров • 4 месяцев назад

Today we release Rhaister, an elegant statistical model that predicts drug phenotypes in new contexts w/ accuracies comparable to experimental assays. And dropping Emerald Bay, a 2M cell dataset measuring long time-course phenotypes across 1000s of drug-cell line interactions.
0:10

Today we release Rhaister, an elegant statistical model that predicts drug phenotypes in new contexts w/ accuracies comparable to experimental assays. And dropping Emerald Bay, a 2M cell dataset measuring long time-course phenotypes across 1000s of drug-cell line interactions.

Nima Alidoust

63,283 просмотров • 2 месяцев назад

Kinda crazy we can now generate realistic-looking 3d models with AI, generated this one in Omma AI and built a #threejs scene around it as if it was a furniture website showcase. Live link: Here is an example prompt that I used:
0:18

Kinda crazy we can now generate realistic-looking 3d models with AI, generated this one in Omma AI and built a #threejs scene around it as if it was a furniture website showcase. Live link: Here is an example prompt that I used:

Marco Fugaro

48,896 просмотров • 3 месяцев назад

Trump still won't cop to America bombing an elementary school in Iran and killing children: "So that's under study right now and we'll give you a report as soon as we have it."
0:15

Trump still won't cop to America bombing an elementary school in Iran and killing children: "So that's under study right now and we'll give you a report as soon as we have it."

Aaron Rupar

48,246 просмотров • 3 месяцев назад

We just crossed another line with AI. This is a fully simulated multiplayer GoldenEye match generated in real-time by a world model. No game engine. No scripted physics. No maps. Just AI generating the world as you play. Insane.
0:11

We just crossed another line with AI. This is a fully simulated multiplayer GoldenEye match generated in real-time by a world model. No game engine. No scripted physics. No maps. Just AI generating the world as you play. Insane.

angel

22,355 просмотров • 3 месяцев назад

1/ Today, we’re announcing our vision for Uniswap v4 🦄 We see Uniswap as core financial infrastructure & think it should be built in public with space for community feedback and contribution. An early implementation of the code can be found here:
0:17

1/ Today, we’re announcing our vision for Uniswap v4 🦄 We see Uniswap as core financial infrastructure & think it should be built in public with space for community feedback and contribution. An early implementation of the code can be found here:

Uniswap

1,762,270 просмотров • 3 лет назад

🌠Today, we’re excited to relaunch Airtable as the AI-native app platform, combining the magic of vibe coding business apps with real production-readiness and scalability, and embedding them with an army of agents that automate thousands of hours of work in seconds. Instead of just adding more AI capabilities to our existing platform, we treated this as a refounding moment for the company. We started with a clean-slate imagining of the ideal form factor for building apps in the agentic era. (If you want to skip all the backstory and just try it out, you can just go to All new signups get the new AI experience, and existing accounts can switch over using this link: Thread and demos below👇
0:18

🌠Today, we’re excited to relaunch Airtable as the AI-native app platform, combining the magic of vibe coding business apps with real production-readiness and scalability, and embedding them with an army of agents that automate thousands of hours of work in seconds. Instead of just adding more AI capabilities to our existing platform, we treated this as a refounding moment for the company. We started with a clean-slate imagining of the ideal form factor for building apps in the agentic era. (If you want to skip all the backstory and just try it out, you can just go to All new signups get the new AI experience, and existing accounts can switch over using this link: Thread and demos below👇

Howie Liu

17,156,319 просмотров • 1 год назад

A viral paper "Language Model Represents Space and Time" recently claims that LLMs learn "world models". As much as I like Max Tegmark's works, I disagree with their definition of world model. World model is a core concept in AI agent and decision making. It is our mental simulation of how the world works given interventions (or lack thereof). A world model captures causality and intuitive physics, telling the agent what is likely and what is impossible. It can and should be used for counterfactual reasoning, i.e. "what ifs": what would happen if I knock over a cup of water? Where would I have been if I had not taken that bus? Yann LeCun Yann LeCun says it well in his position paper ( I quote: "Using such world models, animals can learn new skills with very few trials. They can predict the consequences of their actions, they can reason, plan, explore, and imagine new solutions to problems. Importantly, they can also avoid making dangerous mistakes when facing an unknown situation." The first use of the term World Model in deep policy learning is attributed to hardmaru & Jürgen Schmidhuber: In their seminal paper, an agent masters shooting skills in the popular game Doom (demo below) by learning in imagination, using an internal world model as a "physics simulator". To put in a simple Python math formula, world model learns a function F(s[0:t-1], a) -> s[t:], which takes as input the observed past and current action, and outputs plausible future states. Now the definition of World Model in Tegmark's paper seems to be about predicting GPS coordinates and time eras. I see this as just a classification task with no causal learning and simulation going on. You cannot make meaningful interventions against that model, nor can you optimize any decision making in a closed feedback loop. As for the "space & time neurons", I think they are most similar to the "sentiment neuron" that OpenAI published in 2017: Predicting GPS is conceptually no different from predicting sentiment in my opinion. I don't think their experimental results are wrong - just that their conclusion is on shaky grounds. I welcome any debate! Paper link:
0:15

A viral paper "Language Model Represents Space and Time" recently claims that LLMs learn "world models". As much as I like Max Tegmark's works, I disagree with their definition of world model. World model is a core concept in AI agent and decision making. It is our mental simulation of how the world works given interventions (or lack thereof). A world model captures causality and intuitive physics, telling the agent what is likely and what is impossible. It can and should be used for counterfactual reasoning, i.e. "what ifs": what would happen if I knock over a cup of water? Where would I have been if I had not taken that bus? Yann LeCun Yann LeCun says it well in his position paper ( I quote: "Using such world models, animals can learn new skills with very few trials. They can predict the consequences of their actions, they can reason, plan, explore, and imagine new solutions to problems. Importantly, they can also avoid making dangerous mistakes when facing an unknown situation." The first use of the term World Model in deep policy learning is attributed to hardmaru & Jürgen Schmidhuber: In their seminal paper, an agent masters shooting skills in the popular game Doom (demo below) by learning in imagination, using an internal world model as a "physics simulator". To put in a simple Python math formula, world model learns a function F(s[0:t-1], a) -> s[t:], which takes as input the observed past and current action, and outputs plausible future states. Now the definition of World Model in Tegmark's paper seems to be about predicting GPS coordinates and time eras. I see this as just a classification task with no causal learning and simulation going on. You cannot make meaningful interventions against that model, nor can you optimize any decision making in a closed feedback loop. As for the "space & time neurons", I think they are most similar to the "sentiment neuron" that OpenAI published in 2017: Predicting GPS is conceptually no different from predicting sentiment in my opinion. I don't think their experimental results are wrong - just that their conclusion is on shaky grounds. I welcome any debate! Paper link:

Jim Fan

594,014 просмотров • 2 лет назад

Q. You said that you always had, like, a anger in you. Where does that come from? 🐱 When I was younger, I guess I had a bit of an inferiority complex. As well as some frustration with society in general, of course. I think it was some combination of those things.
0:17

Q. You said that you always had, like, a anger in you. Where does that come from? 🐱 When I was younger, I guess I had a bit of an inferiority complex. As well as some frustration with society in general, of course. I think it was some combination of those things.

슙

69,814 просмотров • 4 месяцев назад

made an app that guesses where you are in the world with just a picture using image embeddings trained on street view data first time using swiftui, consumer apps in general, TestFlight below
0:18

made an app that guesses where you are in the world with just a picture using image embeddings trained on street view data first time using swiftui, consumer apps in general, TestFlight below

Surya Dantuluri

9,602,721 просмотров • 2 лет назад

Russian state TV: “Well, what can be said? Now we understand who runs the world, right? You can say whatever you want about Trump, but he’s a dangerous man. And his victory in 2024 — I still deeply regret that the shooter in Butler didn’t aim an inch to the right”.
0:19

Russian state TV: “Well, what can be said? Now we understand who runs the world, right? You can say whatever you want about Trump, but he’s a dangerous man. And his victory in 2024 — I still deeply regret that the shooter in Butler didn’t aim an inch to the right”.

Serhii Sternenko

135,355 просмотров • 6 месяцев назад