Context compression finally works in production: new research cuts LLM input 16x without the accuracy hit

LCLMs compress LLM context before decode — 8.8x faster at 16x compression, beating every KV cache method tested. Open-sourced by NYU and Columbia.
TVBEurope

BirdDog first to achieve full NDI 6.3 compatibility across product portfolio

Aiming to simplify the deployment of IP video across multi-subnet networks, achieving compatibility reduces manual effort by ...
Memeburn

Google's Gemma 4 12B Runs AI Natively on Your Laptop — No Cloud Needed

Google's Gemma 4 12B brings multimodal AI — audio, video, and text — to a standard 16GB laptop in 2026. No cloud required. Here's what it does and why it matters.
Inavate

Wyrestorm provides AV backbone for University of Strathclyde project

A project at the University of Strathclyde in Glasgow has seen WyreStorm’s NetworkHD AVoverIP ecosystem, delivered in ...
Tech Times

Google Gemma 4 12B Brings Multimodal AI to 16GB Laptops, Free Under Apache 2.0

Google Gemma 4 12B, released June 3, is an open-weight multimodal model that processes text, images, audio, and video in a ...
VentureBeat

Microsoft launches 3 new AI models in direct shot at OpenAI and Google

Microsoft on Thursday launched three new foundational AI models it built entirely in-house — a state-of-the-art speech transcription system, a voice generation engine, and an upgraded image creator — ...
ascopubs.org

Foundation Model Based on Routine Magnetic Resonance Imaging for Brain Tumor Molecular Profiling and Progression Prediction

To build a self-supervised magnetic resonance imaging (MRI) foundation model from routine clinical scans and to test whether it can support key glioma-related applications, including post-therapy ...
marktechpost

Google Introduces T5Gemma 2: Encoder Decoder Models with Multimodal Inputs via SigLIP and 128K Context

T5Gemma 2 follows the same adaptation idea introduced in T5Gemma, initialize an encoder-decoder model from a decoder-only checkpoint, then adapt with UL2. In the above figure the research team show ...
IEEE

An Encoder–Decoder Model Based on Spiking Neural Networks for Address Event Representation Object Recognition

Abstract: Address event representation (AER) object recognition task has attracted extensive attention in neuromorphic vision processing. The spike-based and event-driven computation inherent in the ...