Machine Learning

Auto Added by WPeMatico

Mistral AI Releases Mistral Large 4 (Le Chonk): A 1.05T Parameter Multimodal MoE Model

agentic ai, ai, AI (Artificial Intelligence), AI Infrastructure, AI Shorts, Applications, Artificial Intelligence, Editors Pick, For Devs, Language Model, Large Language Model, Machine Learning, New Releases, Software engineering, Staff, Tech News, Technology

Mistral AI has just announced the release of Mistral Large 4 (ML4), internally nicknamed Le Chonk, as a public preview. ML4 is a granular Mixture of Experts model with 1.05 trillion total parameters, 49 billion active per token, a 1.6 billion parameter vision encoder, and a 1 million token context window, per the model documentation.

Mistral AI Releases Mistral Large 4 (Le Chonk): A 1.05T Parameter Multimodal MoE Model Read More »

Beyond Domain-Specific World Models: JEPA-Anything Uses 1 Recipe for 7 Fields

agentic ai, ai, AI (Artificial Intelligence), AI Infrastructure, AI Paper Summary, AI Shorts, Applications, Artificial Intelligence, Editors Pick, Language Model, Large Language Model, Machine Learning, New Releases, physical ai, Staff, Tech News, Technology

Researchers from PhAI Labs, CUHK, Fudan, Stanford, Oxford and Princeton have released JEPA-Anything, a domain-agnostic framework for building world models. Instead of designing a new predictive model for each field, it applies one shared learning recipe to very different systems. It extends joint-embedding predictive architectures (JEPAs) with a method called Orthogonal Predictive Factorization (OPF). The

Beyond Domain-Specific World Models: JEPA-Anything Uses 1 Recipe for 7 Fields Read More »

Yandex Introduces Sona: A Single Generative Recommender That Replaces Entire Recommendation Cascade

agentic ai, ai, AI (Artificial Intelligence), AI Shorts, Artificial Intelligence, Editors Pick, For Devs, Language Model, Large Language Model, Machine Learning, New Releases, Staff, Tech News, Technology

Most production recommenders are cascades. Candidate generators feed a pre-ranker, which feeds a heavy ranker built on hundreds of engineered features. Yandex’s Sona Technical Report describes a different design. Sona is a generative AI model that brings candidate generation and ranking into a single system, replacing the multiple stages typically used in recommendation pipelines. Yandex

Yandex Introduces Sona: A Single Generative Recommender That Replaces Entire Recommendation Cascade Read More »

The Story of Qwen: Alibaba’s AI Models From 7B to 2.4T

agentic ai, ai, AI (Artificial Intelligence), AI Infrastructure, AI Paper Summary, AI Shorts, Applications, Artificial Intelligence, Decision Models, Editors Pick, Embedding Model, For Devs, Language Model, Machine Learning, New Releases, Small Language Model, Software engineering, Staff, Tech News, Technology, Vision Language Model

In April 2023, Alibaba Cloud demoed a chatbot whose name roughly means ‘truth from a thousand questions.’ Three and a half years later, its descendant ships open weights with 2.4 trillion parameters. This is the story of how Qwen got there, release by release. window.addEventListener(‘message’,function(e){if(e.data&&e.data.mtpQwenTl&&e.data.h){var f=document.getElementById(‘mtp-qwen-tl-frame’);if(f&&e.source===f.contentWindow){f.style.height=e.data.h+’px’;}}}); Chapter 1 — 2023: a thousand questions Alibaba moved

The Story of Qwen: Alibaba’s AI Models From 7B to 2.4T Read More »

Can an Open Model Do Security Research? Cantina’s apex-flash-1 Solves 40 of 60 Held-Out Bug Tasks

agentic ai, ai, AI (Artificial Intelligence), AI Shorts, Applications, Artificial Intelligence, deep learning, Editors Pick, Language Model, Large Language Model, Machine Learning, New Releases, Open Source, Security, Software engineering, Staff, Tech News, Technology

Cantina Security, with Yeta Labs, has released apex-flash-1, an open-weights model trained specifically for vulnerability research. It is a reinforcement learning fine-tune of Z.ai’s GLM-5.3-Flash, released on Hugging Face under the MIT license. Is it deployable? Yes, the MIT weights serve on vLLM, SGLang or Transformers, but BF16 needs roughly 640 GB of GPU memory.

Can an Open Model Do Security Research? Cantina’s apex-flash-1 Solves 40 of 60 Held-Out Bug Tasks Read More »

Google Research Moves Federated Learning Into TEEs: Gboard Now Trains With Externally Verifiable Differential Privacy

ai, AI (Artificial Intelligence), AI Infrastructure, AI Shorts, Applications, Artificial Intelligence, Editors Pick, Federated Learning, Machine Learning, New Releases, Staff, Tech News, Technology

Google Research has announced a next-generation Federated Learning (FL) system built on Trusted Execution Environments (TEEs). The research team claims externally verifiable central differential privacy (DP) guarantees for FL for the first time. What Problem Does TEE-Based Federated Learning Solve? Google introduced Federated Learning FL in 2017. It powers next-word prediction and Smart Compose on

Google Research Moves Federated Learning Into TEEs: Gboard Now Trains With Externally Verifiable Differential Privacy Read More »

Aleph Alpha Releases Kolibri: A 78.1B Open-Weight English-German MoE Model With Only 3.46B Active Parameters

agentic ai, ai, AI (Artificial Intelligence), AI Infrastructure, AI Paper Summary, AI Shorts, Applications, Artificial Intelligence, Editors Pick, For Devs, Language Model, Large Language Model, Machine Learning, New Releases, Open Source, Software engineering, Staff, Tech News, Technology, Uncategorized

Aleph Alpha has released Kolibri, an open-weight Mixture-of-Experts (MoE) language model built for German and English. Kolibri has 78.1B total parameters but activates only 3.46B, or 4.4%, per token. It accepts up to 1,048,576 tokens of context, lets users set reasoning effort per request, and ships under the Apache 2.0 license on Hugging Face. The

Aleph Alpha Releases Kolibri: A 78.1B Open-Weight English-German MoE Model With Only 3.46B Active Parameters Read More »

IBM Brings Bob to Self-Hosted and Air-Gapped Environments: Agentic Software Development Without Moving Your Code

agentic ai, ai, AI (Artificial Intelligence), AI Infrastructure, AI Shorts, Artificial Intelligence, Editors Pick, For Devs, Language Model, Large Language Model, Machine Learning, New Releases, Software engineering, Staff, Tech News, Technology, Uncategorized

IBM has made a self-hosted deployment option for IBM Bob generally available. Bob is IBM’s agentic software development platform. It covers the full lifecycle: understanding code, planning work, executing changes and validating results. The new option lets enterprises run Bob on premises, in private or sovereign clouds, and in air-gapped networks. Is it deployable today?

IBM Brings Bob to Self-Hosted and Air-Gapped Environments: Agentic Software Development Without Moving Your Code Read More »

Prime Intellect Launches Prime Inference: Serverless and Reserved Serving for Frontier Open Models

agentic ai, ai, AI (Artificial Intelligence), AI Infrastructure, AI Shorts, Artificial Intelligence, deep learning, Editors Pick, For Devs, Machine Learning, New Releases, Staff, Tech News, Technology

Prime Intellect has launched Prime Inference, a serving platform for frontier open-source models. It offers serverless endpoints and reserved capacity on Prime’s own GPUs across multiple datacenters. Before public release, it processed nearly a trillion tokens per day internally. That traffic came from RL rollouts, synthetic data generation, evaluations and long-running coding agents. What is

Prime Intellect Launches Prime Inference: Serverless and Reserved Serving for Frontier Open Models Read More »

Computational tools for society’s most complex challenges

ai, AI (Artificial Intelligence), Algorithms, Artificial Intelligence, Autonomous vehicles, Civil and environmental engineering, Computer science and technology, Electrical engineering and computer science (EECS), Faculty, IDSS, Laboratory for Information and Decision Systems (LIDS), Logistics, Machine Learning, MIT Schwarzman College of Computing, Profile, Robotics, School of Engineering, Supply chains, Technology and society, Traffic management, Transportation

As far back as she can remember, Cathy Wu ’12, MNG ’13 wanted to find ways to solve problems to improve people’s lives. Her parents were Taiwanese immigrants, and her father had a long commute to his job, which took him away from the family. On a tight budget, the rest of the family often

Computational tools for society’s most complex challenges Read More »