正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Introducing Rothko, a foundation model of human behaviour designed from the ground up to simulate the world’s most consequential decisions. Rothko is the result of over three years of frontier research at the intersection of behavioural science and machine learning. We’re only just starting to uncover the model’s value... show more

Artificial Societies

3,168 subscribers

21,707 次观看 • 18 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

Today we introduce Liquid Labs, our advanced research unit, with the goal of understanding and building efficient and adaptive intelligence systems. Liquid Labs consolidates our existing research efforts at Liquid across architecture of foundation models, multimodality, training, data, and inference. The lab also will be home to new frontier research work across the broad range of foundation model build-up stack. Read the full announcement: We are hiring: Also find us at NeurIPS 2025 exhibition hall! 🚀

Today we introduce Liquid Labs, our advanced research unit, with the goal of understanding and building efficient and adaptive intelligence systems. Liquid Labs consolidates our existing research efforts at Liquid across architecture of foundation models, multimodality, training, data, and inference. The lab also will be home to new frontier research work across the broad range of foundation model build-up stack. Read the full announcement: We are hiring: Also find us at NeurIPS 2025 exhibition hall! 🚀

Liquid AI

42,942 次观看 • 11 个月前

Looking for the cutting-edge of AI research? Follow Salesforce AI Research to see how we're transforming enterprise technology through advanced innovations. From world models to agentic systems, discover the future of AI before it hits the market.

Looking for the cutting-edge of AI research? Follow Salesforce AI Research to see how we're transforming enterprise technology through advanced innovations. From world models to agentic systems, discover the future of AI before it hits the market.

Salesforce AI Research

2,447,059 次观看 • 1 年前

The onchain dollar is already a ~$310B market. But there is a larger opportunity sitting somewhere else: tens of trillions of bank deposits worldwide. The race to bring them onchain just started and Prividium leads. Here are the findings from the Blockworks Research research 🧵

The onchain dollar is already a ~$310B market. But there is a larger opportunity sitting somewhere else: tens of trillions of bank deposits worldwide. The race to bring them onchain just started and Prividium leads. Here are the findings from the Blockworks Research research 🧵

ZKsync

86,815 次观看 • 2 个月前

We took a 30B model and split it in two to write tokens in parallel instead of one at a time. Introducing Nemotron-Labs-TwoTower: a diffusion language model from NVIDIA Research adapted from Nemotron-3-Nano-30B-A3B. Here’s how it works: one half holds the context, the other writes the tokens, with both reusing the pretrained model instead of training a new one from scratch. We found it kept 98.7% of the original model’s quality at 2.42× faster generation.

We took a 30B model and split it in two to write tokens in parallel instead of one at a time. Introducing Nemotron-Labs-TwoTower: a diffusion language model from NVIDIA Research adapted from Nemotron-3-Nano-30B-A3B. Here’s how it works: one half holds the context, the other writes the tokens, with both reusing the pretrained model instead of training a new one from scratch. We found it kept 98.7% of the original model’s quality at 2.42× faster generation.

NVIDIA AI

763,589 次观看 • 4 个月前

A research study led by scientists from Dubai Health and MBRU has been featured on the cover of Nature Medicine, one of the world’s most prestigios medical journals. This milestone reflects the remarkable progress of Dubai’s scientific research ecosystem, elevating quality of life and advancing health for humanity. To learn more about the study, visit the link:

A research study led by scientists from Dubai Health and MBRU has been featured on the cover of Nature Medicine, one of the world’s most prestigios medical journals. This milestone reflects the remarkable progress of Dubai’s scientific research ecosystem, elevating quality of life and advancing health for humanity. To learn more about the study, visit the link:

Dubai Health

14,031 次观看 • 6 个月前

🧵How do you reduce research from two years to 6 weeks? With more than 100 partners across hospitals, research groups, and institutions in Asia and the US, we’re focused on building tools that actually reach people sooner. Here’s the impact of Life AI. ⬇️

🧵How do you reduce research from two years to 6 weeks? With more than 100 partners across hospitals, research groups, and institutions in Asia and the US, we’re focused on building tools that actually reach people sooner. Here’s the impact of Life AI. ⬇️

LIFE AI

21,420 次观看 • 10 个月前

Frontier drug discovery programs require frontier capabilities. Today, we’re excited to share that BMS will be using Chai’s models to support the discovery of new therapeutics. Since rolling out Chai-3 to partners at the start of this year, we’ve partnered with Eli Lilly and Company, Novartis, Pfizer Inc., argenx, and now Bristol Myers Squibb. Many of the world’s leading biopharma companies are now using our models to go after diseases. Scientific discovery is the domain where access to frontier AI matters the most. At Chai Discovery, we’ve seen first hand how our latest model generations can make the difference between discovering a molecule and not. We’re excited to work with Bristol Myers Squibb to break new ground in medicine and science.

Frontier drug discovery programs require frontier capabilities. Today, we’re excited to share that BMS will be using Chai’s models to support the discovery of new therapeutics. Since rolling out Chai-3 to partners at the start of this year, we’ve partnered with Eli Lilly and Company, Novartis, Pfizer Inc., argenx, and now Bristol Myers Squibb. Many of the world’s leading biopharma companies are now using our models to go after diseases. Scientific discovery is the domain where access to frontier AI matters the most. At Chai Discovery, we’ve seen first hand how our latest model generations can make the difference between discovering a molecule and not. We’re excited to work with Bristol Myers Squibb to break new ground in medicine and science.

Joshua Meier

82,432 次观看 • 2 个月前

This also shows up in the representations learned by the model. We plot the model’s representations of human and robot images. As pre-training is scaled up, the representation of humans and robots become more aligned: to a scaled-up model, human videos "look" like robot demos.

This also shows up in the representations learned by the model. We plot the model’s representations of human and robot images. As pre-training is scaled up, the representation of humans and robots become more aligned: to a scaled-up model, human videos "look" like robot demos.

Physical Intelligence

121,658 次观看 • 10 个月前

1/ today, with the launch of muse image and the preview of muse video, we wanted to share more samples from the model alongside research details and eval results see some more samples from muse video below (and see thread for research blog!)

1/ today, with the launch of muse image and the preview of muse video, we wanted to share more samples from the model alongside research details and eval results see some more samples from muse video below (and see thread for research blog!)

Alexandr Wang

380,497 次观看 • 4 个月前

Revealing the world’s best science-backed culinary AI soon. Research preview of our 300-dimensional flavour similarity model created using ML.

Revealing the world’s best science-backed culinary AI soon. Research preview of our 300-dimensional flavour similarity model created using ML.

Josef Chen

14,782 次观看 • 1 年前

At Applied Compute, we're not just training models, we're building a model factory. To help researchers scale, we built Ari, our in-house AI research agent. Ari diligently sifts through gigabytes of logs to find evidence of unhealthy behavior, remembers results the team has learned from past experiments, and produces styled research reports for the team to review.

At Applied Compute, we're not just training models, we're building a model factory. To help researchers scale, we built Ari, our in-house AI research agent. Ari diligently sifts through gigabytes of logs to find evidence of unhealthy behavior, remembers results the team has learned from past experiments, and produces styled research reports for the team to review.

Applied Compute

42,116 次观看 • 2 个月前

InterLink’s Journey to the World’s Top 10 Most Accurate AI Models Artificial Intelligence has rapidly become the defining force of this decade powering breakthroughs across every industry. But while most projects chase trends, InterLink Labs 👤 + 🌐 has been quietly building something deeper: an AI ecosystem grounded in verified human intelligence. Long before AI captured global headlines, InterLink Labs 👤 + 🌐 had already begun its research and engineering efforts back in 2019, assembling a world-class team of engineers and researchers from Big Tech companies and top QS-ranked universities. Their vision was clear - to build a model that truly understands humans, not just data. Unlike traditional AI systems trained purely on digital information, InterLink Labs 👤 + 🌐’s Human-AI Model learns from verified human behavior across millions of Human Nodes. This unique layer of authentic, real-world human input gives InterLink’s AI an unprecedented advantage in trustworthiness, bias reduction, and contextual understanding. Beyond algorithmic optimization, InterLink Labs 👤 + 🌐’s R&D efforts are being scaled to an unprecedented level. The team operates over 100 NVIDIA H100 servers, processing massive volumes of verified behavioral data contributed by real Human Nodes across the world. This data - diverse, decentralized, and human-validated forms the foundation of a next-generation intelligence system designed to mirror real human reasoning patterns. At the same time, InterLink Labs 👤 + 🌐’s AI-powered Human Credit Score applies advanced machine learning to analyze authenticity, contribution, and reliability. Creating an ethical model of digital reputation and fairness. Aligned with National Institute of Standards and Technology (National Institute of Standards and Technology) evaluation standards, InterLink Labs 👤 + 🌐 now aims to achieve Top 10 accuracy globally among AI models. Competing with research teams from Samsung Electronics, Kakao, キヤノン株式会社 / Canon Inc., and other global giants, InterLink Labs 👤 + 🌐’s engineers continue to train, benchmark, and refine their architecture daily to reach world-class precision and consistency. But this is more than a technical race. It’s a human mission. Every verified user contributes to the world’s first Human-Powered Intelligence Network, where real people fuel the evolution of trustworthy AI. As InterLink Labs 👤 + 🌐 advances toward global National Institute of Standards and Technology recognition, one truth becomes clear: The future of intelligence won’t be artificial. It will be human-powered. #InterLink #ITLG #ITL

InterLink’s Journey to the World’s Top 10 Most Accurate AI Models Artificial Intelligence has rapidly become the defining force of this decade powering breakthroughs across every industry. But while most projects chase trends, InterLink Labs 👤 + 🌐 has been quietly building something deeper: an AI ecosystem grounded in verified human intelligence. Long before AI captured global headlines, InterLink Labs 👤 + 🌐 had already begun its research and engineering efforts back in 2019, assembling a world-class team of engineers and researchers from Big Tech companies and top QS-ranked universities. Their vision was clear - to build a model that truly understands humans, not just data. Unlike traditional AI systems trained purely on digital information, InterLink Labs 👤 + 🌐’s Human-AI Model learns from verified human behavior across millions of Human Nodes. This unique layer of authentic, real-world human input gives InterLink’s AI an unprecedented advantage in trustworthiness, bias reduction, and contextual understanding. Beyond algorithmic optimization, InterLink Labs 👤 + 🌐’s R&D efforts are being scaled to an unprecedented level. The team operates over 100 NVIDIA H100 servers, processing massive volumes of verified behavioral data contributed by real Human Nodes across the world. This data - diverse, decentralized, and human-validated forms the foundation of a next-generation intelligence system designed to mirror real human reasoning patterns. At the same time, InterLink Labs 👤 + 🌐’s AI-powered Human Credit Score applies advanced machine learning to analyze authenticity, contribution, and reliability. Creating an ethical model of digital reputation and fairness. Aligned with National Institute of Standards and Technology (National Institute of Standards and Technology) evaluation standards, InterLink Labs 👤 + 🌐 now aims to achieve Top 10 accuracy globally among AI models. Competing with research teams from Samsung Electronics, Kakao, キヤノン株式会社 / Canon Inc., and other global giants, InterLink Labs 👤 + 🌐’s engineers continue to train, benchmark, and refine their architecture daily to reach world-class precision and consistency. But this is more than a technical race. It’s a human mission. Every verified user contributes to the world’s first Human-Powered Intelligence Network, where real people fuel the evolution of trustworthy AI. As InterLink Labs 👤 + 🌐 advances toward global National Institute of Standards and Technology recognition, one truth becomes clear: The future of intelligence won’t be artificial. It will be human-powered. #InterLink #ITLG #ITL

InterLink Labs 👤 + 🌐

51,650 次观看 • 1 年前

Over the past months, Cohort I of our RL Residency has been shipping. Highlights - continual learning - automating AI research (from GPU programming to RL itself) - embodied environments - multi-agent systems - materials science discovery

Over the past months, Cohort I of our RL Residency has been shipping. Highlights - continual learning - automating AI research (from GPU programming to RL itself) - embodied environments - multi-agent systems - materials science discovery

Prime Intellect

59,938 次观看 • 6 个月前

We’re excited to announce this strategic partnership between Litecoin Foundation ⚡️ and DeLorean Labs ⚡️🏎️⚡️ Litecoin and DeLorean will work together on engaging activations and new initiatives at the intersection of automotive innovation and cryptocurrency adoption. We’re also honored to collaborate with Charlie Lee Ⓜ️🕸️, a true pioneer who helped build the crypto and Web3 industry from the ground up. As newcomers to the space, we’re grateful for the chance to learn from one of its most respected builders. 🙏🤝 When this baby hits 88 mph, you’re going to see some serious sh!t! $LTC solana:DmcqxT7WXgXhq2LHURQ2h8GA5PeCBuMTLeyaj6MXhf3p

We’re excited to announce this strategic partnership between Litecoin Foundation ⚡️ and DeLorean Labs ⚡️🏎️⚡️ Litecoin and DeLorean will work together on engaging activations and new initiatives at the intersection of automotive innovation and cryptocurrency adoption. We’re also honored to collaborate with Charlie Lee Ⓜ️🕸️, a true pioneer who helped build the crypto and Web3 industry from the ground up. As newcomers to the space, we’re grateful for the chance to learn from one of its most respected builders. 🙏🤝 When this baby hits 88 mph, you’re going to see some serious sh!t! $LTC solana:DmcqxT7WXgXhq2LHURQ2h8GA5PeCBuMTLeyaj6MXhf3p

DeLorean Labs

58,861 次观看 • 1 个月前

The future of bio is powered by faster data Introducing the Medra AI Experimentalist: an agent that turns goals into experimental designs, learns from every result, and develops the next assay Excited to collaborate with DARPA and NVIDIA Healthcare on the future of science

The future of bio is powered by faster data Introducing the Medra AI Experimentalist: an agent that turns goals into experimental designs, learns from every result, and develops the next assay Excited to collaborate with DARPA and NVIDIA Healthcare on the future of science

Michelle Lee

29,146 次观看 • 4 个月前

Alibaba Cloud's research highlights widespread optimism about AI adoption across Asia. The majority of companies surveyed expect to significantly increase their use of AI applications over the next year. With planned investments and positive sentiment, AI growth is poised to accelerate across the region.🚀

Alibaba Cloud's research highlights widespread optimism about AI adoption across Asia. The majority of companies surveyed expect to significantly increase their use of AI applications over the next year. With planned investments and positive sentiment, AI growth is poised to accelerate across the region.🚀

Alibaba Group

944,483 次观看 • 2 个月前

What happens when AI agents put thousands of GPUs to work improving AI itself? Meet OpenRSI Index. 👇 The goal: measure whether AI research agents can improve on human-designed training recipes. Real model-development tasks: 🔹 Pre-training 🔹 Post-training 🔹 Vision generation ⚡ Up to 20,864 H100-hours per run in the featured examples. Can agents discover better methods? Let’s put them to the test. Great work by OpenRSI Foundation and the research community! 🚀

What happens when AI agents put thousands of GPUs to work improving AI itself? Meet OpenRSI Index. 👇 The goal: measure whether AI research agents can improve on human-designed training recipes. Real model-development tasks: 🔹 Pre-training 🔹 Post-training 🔹 Vision generation ⚡ Up to 20,864 H100-hours per run in the featured examples. Can agents discover better methods? Let’s put them to the test. Great work by OpenRSI Foundation and the research community! 🚀

Pan Lu

45,115 次观看 • 1 个月前

Population dynamics can provide insights into domains ranging from health to environmental science. Here we introduce a geospatial foundation model (plus embeddings and code recipes) that could be employed for a variety of downstream tasks. →

Population dynamics can provide insights into domains ranging from health to environmental science. Here we introduce a geospatial foundation model (plus embeddings and code recipes) that could be employed for a variety of downstream tasks. →

Google AI

41,934 次观看 • 2 年前

We sent a swarm of AI agents to solve Karpathy’s NanoChat benchmark, and they crushed SoTA in 3 days. We built our own harness on top of a graph database to do auto-autoresearch, each iteration learning from mistakes to do research better. The team wrote >15,000 entries. 🧵

We sent a swarm of AI agents to solve Karpathy’s NanoChat benchmark, and they crushed SoTA in 3 days. We built our own harness on top of a graph database to do auto-autoresearch, each iteration learning from mistakes to do research better. The team wrote >15,000 entries. 🧵

Dan Kondratyuk

314,729 次观看 • 1 个月前