Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
A developer rebuilt an over-engineered parity classifier after finding label leakage and a mislabeled unsupervised framing in the original experiment, moving to a stricter train/validation/test protocol over all integers…
Kuaishou's A/B Agent, an AI system that studied 310 historical recommendation strategies and iteratively proposed production configurations, delivered a 4.829% gross merchandise value increase over the production compari…
Developer zatsepin released jigor, a Rust-based zero-shot classifier runner that executes JEV-style decision models locally via ONNX or remotely through OpenRouter's Decisions API. The tool installs via npm (@zatsepin/ji…
A Towards Data Science experiment found that AI detectors flagged many genuine reviews, and filtering the flagged content made the sentiment model less accurate. The author tested three methods for spotting AI slop in tr…
A developer released FLM (Fly Language Model), a language model trained from scratch on a connectome-derived recurrent network built from a modeled subset of fruit-fly brain wiring, with no pretrained transformer supplyi…
BlueVetaUpright, a purpose-built product-photo orientation system, reached 99.02% accuracy on clean cardinal rotations, 96.67% across the full 360° range, and 84.65% on rotations exactly ±40° from any clean angle, accord…
Flytrex is rolling out a predictive fleet positioning system with its Rooftop Dock in Dallas-Fort Worth that uses machine-learning models to send delivery drones to restaurant rooftops before orders are placed, a setup t…
A paper titled "Learning to Solve Hard Problems in RL for LLMs by Never Giving Up" proposes countering the Matthew Effect in reinforcement learning, where LLMs improve far more on easy problems they already solve than on…
A tutorial demonstrates an end-to-end multimodal data augmentation and adversarial robustness workflow using Meta's AugLy library for images, text, and audio, connected directly to PyTorch datasets and DataLoaders. The w…
Developer siritoriyowai released pysimplicial, an experimental Python library for simplicial complexes and topology-oriented machine learning, installable via pip from the kaifczxc-lab GitHub repository. The toolkit prov…
A high-school student built and deployed DiabetesRiskAnalyzer.org, a free open-source Flask web application that predicts diabetes risk from lifestyle markers including blood pressure, BMI and age, and published the unde…
A developer built PipelineSentinel, a deployable layer that scores each CI/CD run against its own pipeline's history to flag anomalies before a human has to triage them. The tool pulls run metadata from the GitHub Action…
A developer has published a technical guide showing how to combine differential privacy and decentralized learning to build HIPAA-ready medical analytics, using PySyft for federated computation and Opacus for differentia…
Xiaomi released MiMo-V2.6-Flash-RL, a sparse Mixture-of-Experts omnimodal model with 309 billion total parameters and 15 billion active per token, positioned as the efficiency-focused sibling to MiMo-V2.6-Pro-RL with a 1…
Edge0 released Audio8 ASR Infinite, an open-weight streaming speech recognition model under the Apache 2.0 license, available on Hugging Face as Edge0/Audio8-ASR-Infinite and on GitHub as Edge0-AI/Audio8-ASR-Infinite. Th…
Francis Bach's blog post analyzes why Monte Carlo estimates of log-sum-exp functions suffer exploding variance, showing that for independent normal variables with mean μ and variance σ², the relative squared error of est…
Qdrant released a public 24-terabyte dataset of 10 billion vectors that provides the first precisely computed ground truth for Internet-scale vector search, after spending massive GPU compute on more than a quadrillion (…
A developer built Stock Price Prediction, an open-source full-stack application that applies seven machine learning algorithms to 1,001 stocks across US and Indian markets, using a Flask, TensorFlow and scikit-learn back…
Stanford and Nvidia released CLM-8B on September 23, an open-weights Contrastive Language Model that makes real-time agent decisions up to 9x faster than TypeSafe AI's proprietary Jev model, clocking 16.5 milliseconds pe…
A developer trained a LeWorldModel JEPA-based world model locally on an RTX 3080 Ti to play Pokémon Red, and the model learned to plan a sequence of button presses that selected the Squirtle starter from a saved state in…