Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
Open-weight AI models are now 3–6 months behind frontier cloud models in benchmark performance, but the gap is closing fast enough that local AI has become a viable infrastructure decision for cost, privacy, and agentic …
HASH researchers argue that building a process foundation model (PFM) requires a novel architecture, Dynamic Graph-JEPA, which extends Graph-JEPA to handle partially-observed, typed, temporal graph state. They claim exis…
Hugging Face researchers released a new method called delta weight sync that reduces the per-step weight transfer in asynchronous reinforcement learning from 1.2 GB to as little as 20 MB for a 0.6B parameter model. The t…
A team of researchers from Yale's Building AI Infrastructure class achieved a new speed record of 81.2 seconds on the modded-nanoGPT benchmark by introducing a learnable per-(layer, head) tanh(α) gate on Exclusive Self-A…
Chan Zuckerberg Biohub released ESMC, ESMFold2, and ESM Atlas under an MIT license this morning, and Sheaf v0.11 shipped support for both model backends within twelve hours. The same-day integration was possible because …
Ollama and LM Studio now allow users to run open-weight AI models like Qwen 3, Gemma 3, and DeepSeek-R1 locally on consumer hardware without API keys or monthly fees. The tools support quantized models in GGUF format, en…
A developer has released one of the first publicly documented LoRA training runs and general-purpose visual-enhancement LoRAs for HiDream-O1-Image, a top-ranked open-weight text-to-image model that shipped inference-only…
Hussein Mahdi published a breakdown of the mathematics underlying neural networks, aiming to clarify concepts he felt were poorly explained elsewhere. The guide walks through the core calculations that power deep learnin…
0G Labs, in partnership with China Mobile, successfully trained a 107-billion-parameter AI model using its DiLoCoX framework, marking the first decentralized training of a model exceeding 100 billion parameters. The proj…
Anthropic's Claude Opus 4.7 is positioned as a top-tier frontier model for writing and coding, while competitors like OpenAI's GPT-5.5, Google's Gemini 3.1 Pro, and DeepSeek V4 each target different strengths from genera…
Xiangyu Chen is a Senior Research Engineer in the Autonomous Vehicle Research Group at NVIDIA, where he focuses on translating heterogeneous multimodal data into learning objectives for robust autonomous systems. Before …
Jason Stock, a researcher specializing in artificial intelligence, machine learning, computer vision, and generative AI, focuses his work on climate simulation. His contributions to the field are documented through his G…
Researchers have found that large language models exhibit severe bias and mode collapse when asked to generate random outputs, with models like Qwen3 selecting "Wednesday" 80% of the time when asked for a random weekday …
NVIDIA released CompileIQ, an AI-powered compiler auto-tuning framework that uses evolutionary and genetic algorithms to optimize GPU compilers for individual workloads. The tool, included in NVIDIA CUDA 13.3, generates …
NVIDIA released CUDA 13.3, introducing tile programming in C++ that automates low-level GPU management for optimized kernel development across all supported architectures. The update also includes CUDA Python 1.0 with gr…
A developer implemented FlashAttention's forward and backward passes from scratch in pure CUDA C++, achieving O(N) memory complexity through manual SRAM tiling and online softmax recurrence. A rejected `llama.cpp` pull r…
Ninety-seven percent of telecom executives are assessing or adopting AI to improve customer experiences and network operations, according to NVIDIA's 2025 State of AI in Telecommunications report. Despite generating posi…
Apollo Research researchers have identified practical refinements to Synthetic Document Finetuning (SDF), a knowledge editing technique that implants beliefs into AI models by training them on LLM-generated documents. Th…
Amazon Web Services announced a new solution for building highly scalable, serverless multi-agent generative AI systems using LangGraph agents integrated with Amazon Bedrock AgentCore Memory and Observability. The approa…
NVIDIA, Amazon Web Services, and Strands have launched a multi-agent generative AI system that combines NVIDIA NIM for GPU-accelerated inference, Amazon Bedrock AgentCore for managed runtime and shared memory, and Strand…