Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
Researchers have developed MobileGym, an Android-like operating system that runs entirely in a browser, enabling verifiable and highly parallel simulation for mobile GUI agent research. The platform covers 28 real-world …
A new formal proof submitted to arXiv on May 25, 2026, by David Klindt, Yann LeCun, and Randall Balestriero establishes the precise mathematical conditions under which LeJEPA, a variant of the Joint Embedding Predictive …
A developer created a Linux kernel module and userspace shim that transforms a standard USB4 connection into a low-latency InfiniBand device, enabling distributed AI inference across two 128GB Strix Halo mini PCs. The pr…
Ashwin Gururaj, a data scientist and AI engineer based in Melbourne, Australia, is seeking full-time, contract, or internship opportunities. He recently built Sift, an open-source multi-agent fact-checking pipeline that …
Trelk, a new app for reading and knowledge management, launched with AI-powered features that automatically extract, summarize, tag, and connect articles, papers, and notes into a personal knowledge graph. The app runs e…
Jack Maguire, a writer and analyst, developed a strict 100-point data rubric to eliminate streaming service decision fatigue by scoring movies on character credibility, dialogue, genre fit, and resonance, with defensive …
A developer has created a system called "skill distillation" that uses frontier AI models like Opus 4.7 and GPT-5.1 to write procedural skill files, which are then executed by smaller local models like Qwen 35B or Gemma …
Together AI built the world’s fastest speech-to-text stack, enabling NVIDIA’s Parakeet-TDT 0.6B v3 model to transcribe roughly 20 hours of speech in under 10 seconds. The company achieved this by optimizing the entire da…
PyTorch released a beginner's guide to its torch.profiler tool, starting with profiling a simple matrix multiplication and addition operation on an A100 GPU. The guide walks through reading profiler traces and tables to …
VDF AI built SEEMR, a self-evolving model router, to replace static rule tables for enterprise AI workload routing because rule tables cannot adapt to changing model catalogs, workload mixes, and runtime behavior. The ro…
Josh Engroff, a Python developer, built a Rust-based JSON parser that outperforms CPython's C-based stdlib json module by up to 2.96× after completing a 6-week Python-to-Rust cohort. His streaming parser, which uses zero…
Researchers developed and validated a large language model-based linguistic analysis to predict Alzheimer disease, framing early detection as critical given the condition's progressive nature and rising global prevalence…
Serenity, an autonomous AI agent built with a memory system inspired by biological neural processes rather than traditional databases, can now run entirely on local consumer hardware without cloud connectivity or API key…
Liquid AI released LFM2.5-8B-A1B, an on-device Mixture-of-Experts model with 8.3 billion total parameters and 1.5 billion active parameters per token, designed for tool calling and reasoning on consumer hardware. The mod…
Roblox researchers have released CubePart, an open-source generative AI framework that creates 3D objects with separately controllable, labeled parts from text prompts and user-defined part lists. The two-stage diffusion…
Researchers found that superhuman Go AIs remain vulnerable to adversarial attacks despite implementing multiple defensive countermeasures, including adversarial training and architectural changes. None of the tested defe…
Apple will present 14 AI research papers at the 2026 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) in Denver from June 3-7, covering image generation, spatial understanding, multimodal reasoning, …
An international research team found that large language models integrated false statements into their outputs even when those claims were explicitly labeled as false in training data. The researchers seeded fine-tuning …
A solo developer launched Agenlus, a browser-based reinforcement learning training platform designed to make RL accessible without requiring expensive hardware or installations. The platform allows users to train agents,…
Azercell Telecom LLC, Azerbaijan's leading telecommunications provider, trained an Azerbaijani large language model on Amazon SageMaker AI in collaboration with the AWS Generative AI Innovation Center. The six-week proje…