Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
An international team including researchers from the University of Manchester, University College Dublin, the European Space Agency, Diamond Light Source and South East Technological University captured metal dendrites f…
Towards Data Science published "Your Model's MSE Is Lying to You," the first post in a series on probabilistic forecasting for physical signals, with a follow-up planned on rolling a forecast forward more than one step. …
Reward AI, a Stanford robotics spinout founded by Zipeng Fu and Chen Wang, launched OM-1 on September 14th, a robot foundation model the company says controls tabletop arms, industrial systems and humanoids without robot…
Engineers at the USC Viterbi School of Engineering and USC Stevens identified where AI models misinterpret emotional cues from audio and video and introduced a framework to help them reason with real evidence, according …
Transolver, UPT, and AB-UPT each replace full point-to-point self-attention with smaller token sets to make global computation affordable on industrial-scale meshes, according to Part 5 of the Geometric Deep Learning ser…
NVIDIA Transformer Engine is now integrated with JAX to accelerate dropless Mixture of Experts (MoE) training, targeting the conditional computation bottleneck in architectures used by DeepSeek, Qwen, and Mixtral. The in…
NVIDIA reported that combining its Transformer Engine library with the JAX Python library raised DeepSeek-V3 MoE training throughput on NVIDIA GB200 GPUs from an unoptimized baseline of 103 TFLOPS/GPU to 1,068 TFLOPS/GPU…
A developer released neural_texture_es2, a self-contained C++ experiment that encodes an RGB image or up to four PBR material textures as a shared low-resolution latent texture plus a tiny MLP decoder, trained entirely w…
Researchers Florian Hahlbohm, Jorge Condor, Linus Franke, Martin Eisemann and Marcus Magnor published an arXiv paper introducing a compact neural appearance model for 3D Gaussian Splatting that reduces the per-primitive …
Researchers at MIT developed an algorithm called HardFlow that forced generative AI outputs to satisfy strict, non-negotiable safety constraints in simulated robotics and imaging tests, meeting the required rule every ti…
Google's BigQuery added six augmented analytics Table-Valued Functions that combine AI, ML and statistical methods to automate insight discovery directly where data lives. The functions — AI.KEY_DRIVERS, AI.CAUSAL_EFFECT…
AWS published a technical walkthrough for an unattended retail replenishment loop that combines Databricks Many Model Forecasting (MMF) serving Chronos-2, a Databricks Genie Agent, and Amazon Quick Flows. The system fore…
The IEEE P802.3dj Task Force is finalizing the 1.6 terabit Ethernet standard, with ratification expected later this year, as AI-driven demand for data center bandwidth accelerates. Industry groups are demonstrating inter…
A developer published an introductory guide to AI engineering explaining that traditional deterministic software fails on tasks like detecting customer frustration, while large language models work as prediction machines…
Researchers introduced Simple Attention Sparsification (SAS), a gated sparse attention mechanism that optimizes context ranking end-to-end with the language modeling loss by injecting the selector's continuous scores int…
A technical handbook on FlashAttention explains that the algorithm achieves wall-clock speedups by reducing data movement between GPU high-bandwidth memory (HBM) and on-chip SRAM rather than by approximating attention, u…
A prospective PhD student planning to study Graph ML, spanning graph signal processing and graph deep learning, is seeking advice on branding their doctoral research to land a research scientist role at big tech or a com…
A developer released a from-scratch Sudoku solver built on a Looped MLP-Mixer architecture, trained on the sapientinc/sudoku-extreme dataset and runnable on an NVIDIA Jetson Nano via JetPack Docker. The model reuses a sh…
Model distillation, a technique in which a smaller "student" model learns to replicate the behavior of a much larger "teacher" model, has become one of the most contested topics in the AI industry as unauthorized distill…
An AI system trained on more than 20,000 protein structures from pharmaceutical companies outperforms AlphaFold-like models that rely only on public data, according to a Nature report by Ewen Callaway published September…