Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
An SRE's exploration of support vector machines highlights the kernel trick as a classic engineering move: transforming problems into spaces where they become trivial. The post emphasizes that SVM's robustness comes from…
A study by Łucki et al. found that machine unlearning methods are vulnerable to knowledge recovery, but experiments on six unlearned Llama-3-8B-Instruct checkpoints showed that ablating the refusal direction in the ILU-R…
Portail, a sovereign Rust AI gateway, has released a BitNet B1.58 SIMD tensor engine achieving 16.9k tokens per second with 8.0x FP16 compression, enabling real-time sliding window response latencies across 8 fleet nodes…
A developer who previously worked on a reverse-engineered Apple Neural Engine (ANE) driver for Linux has resumed the project, aiming to map the full internal architecture of the M1 ANE, three years after abandoning it du…
Google's Pixel phones have used AI and machine learning for over a decade to advance phone photography, with features like Night Sight, Magic Eraser, and Best Take, according to an exclusive interview with Isaac Reynolds…
A developer has built an early warning system that uses LSTM networks to detect flu and infections by analyzing resting heart rate data from wearable devices. The system, optimized for Edge AI deployment via TensorFlow L…
Rallylens, a web app built by a student and tennis player, uses AI-powered video analysis to provide real-time stroke feedback without a coach. The system segments uploaded videos frame by frame, identifies stroke types …
A developer benchmarked Google's Gemma 4 E2B model on a Cloud TPU v6e-1 (Trillium) and found that while the chip offers 3.6x more KV cache memory than v5e-1, it costs 2.25x more and delivers only 1.62-1.68x throughput on…
An AWS developer built an image processing pipeline that uses a segmentation model inside a Lambda container to straighten, crop, and remove backgrounds from photos of AWS Builder Cards. The model, BiRefNet-General-Lite,…
Makerfabs released the Nomad Terminal, a handheld development platform built around an ESP32-S3-WROOM-1 module with a dual-core Tensilica LX7 CPU up to 240 MHz, a Semtech LR1121 LoRa transceiver, and a 3.5-inch 480×320 T…
Gartner, Inc. is hiring a Software Engineer in Gurgaon with 2-4 years of experience in Python, NLP, BERT, and generative AI to build AI/ML applications for its Consulting & Insights Technology. The role involves developi…
Gartner is hiring an Associate Data Scientist for its Insights & Product Analytics team in Gurgaon, India, to work on high-impact data science projects involving NLP, machine learning, deep learning, and generative AI. T…
Weaviate's new `effort` parameter in Query Agent Search Mode, available in `weaviate-agents 1.8.0` and `agents-typescript-client 1.7.0`, lets users scale test-time compute per request, lifting nDCG@10 to 57.5 on BRIGHT B…
DeepGrove released Maple-Preview, an open-weight 20B-A1B ternary-weight mixture-of-experts reasoning model that runs at 218 tokens per second on an Apple Mac mini M4, with a 5.31 GB checkpoint and a 131,072-token context…
LiquidAI and collaborators released fuse-1 Lite, a 5.72 billion parameter mixture-of-experts model that transplants 960 coding experts from Qwen3.6-35B-A3B into LiquidAI's LFM2.5-2.6B host model, achieving coding perform…
NVIDIA has released SwitchYard, an open-source routing library that automatically directs AI agent tasks to the most appropriate local or frontier model, claiming it can deliver 50% faster responses and 25% better token …
Fuse-1 Lite, a 5.72B parameter mixture-of-experts coding model from LiquidAI, can run locally with VRAM needs ranging from 3.36 GB in 4-bit quantized form to about 12 GB in full bfloat16 precision, according to its model…
Motley AI's open-source semantic layer SLayer improved text-to-SQL accuracy by over 34% on production-scale databases, with the SLayer agent passing 153 of 265 tasks (57.7%) versus 114 (43.0%) for a raw-SQL agent on Live…
A developer's third installment in a GPU optimization series details implementing distributed training for large language models using CUDA and ROCm, covering All-Reduce, Ring-AllReduce, and ZeRO sharding with NCCL and R…
CATL, the world's largest battery maker, announced that its aviation battery system for passenger eVTOL aircraft passed a key safety test, becoming the world's first aviation battery system using prismatic cells with an …