Hardware
The silicon and systems behind the AI boom: GPU research, custom accelerator papers, inference optimization techniques, quantization methods, and the hardware engineering that determines training economics and deployment feasibility.
Robots Follow Hand-Drawn Paths in Real Space
A new mixed reality system lets users draw navigation routes directly on the floor, improving accuracy and ease of use for controlling mobile robots in shared environments.
AI Speeds Up Short Conversations on Powerful GPUs
A simple tweak to how AI models process short prompts can cut response times by over 20%, making chatbots faster without changing the underlying technology.
AI Compression Tool Makes Large Models Fit on Consumer GPUs
A new open-source framework automates the complex process of shrinking AI models, allowing organizations to deploy powerful language models without expensive hardware upgrades.
AI Training Gets a Time-Saving Rulebook
A new study reveals how to pick the right model size when you're racing against the clock on consumer hardware, showing that bigger isn't always better under time pressure.
AI Guides Viewers in VR Without Breaking Immersion
A new system uses natural language and visual effects to direct attention in 360-degree videos, overcoming limitations of previous methods that left users lost in virtual environments.
New Method Speeds Up Circuit Simulation Dramatically
Researchers combine parallel processing and system reduction to cut simulation times by up to 97%, enabling faster design of electronics from smartphones to medical devices.
Robots Master Curved Surfaces with New Two-Stage Method
A novel AI framework enables robots to execute precise, repetitive tasks on complex curved surfaces without drifting or losing contact, using minimal sensors and offline geometric adaptation.
Robots Learn to Climb Stairs with Less Power
A new AI method teaches hybrid robots to blend flying and driving, cutting energy use by 38% on real-world stair-climbing tasks.
Ray Tracing Cores Find New Life Beyond Gaming
A comprehensive review reveals how specialized GPU hardware, designed for realistic lighting in games, can accelerate scientific simulations and data analysis by up to 200 times, offering a new path for high-performance computing.
AI Extends Firefighter Breathing Time by a Third
A new AI-controlled life support system for firefighters uses sensor fusion and predictive optimization to manage limited oxygen and scrubbers, boosting endurance by up to 34% while maintaining safety in extreme heat.
A Universal Language for All GPUs Emerges from Hardware Necessity
Researchers identify ten fundamental computational primitives shared across all major GPU vendors, paving the way for a vendor-neutral instruction set that could break software lock-in and boost portable performance.
AI Serves Large Models Faster with Smarter Replication
A new method allocates AI model replicas based on need, boosting inference speed by up to 20% while using far less memory than current techniques.
New Algorithm Tracks Billions of Particles on Everyday Computers
A breakthrough in simulation software allows scientists to model complex systems like clouds and engines with unprecedented scale and speed, using standard hardware.
AI Spots Silent GPU Failures Before They Crash Systems
A new method detects when GPUs fail without warning by tracking monitoring system breakdowns, offering earlier alerts for high-performance computing and AI workloads.
Brain-Inspired AI Could Slash Edge Device Power Use
A new review reveals how spiking neural networks mimic the brain's efficiency to process data with minimal energy, but hardware mismatches still block widespread adoption.
New Hardware Security Method Defeats AI Attacks
A simple resistor-capacitor device can protect Internet of Things gadgets from machine learning-based hacking attempts, offering a lightweight alternative to complex encryption.
AI Solves Complex Logic Puzzles 8 Times Faster
A new hybrid AI method combines GPU learning with traditional logic solving to dramatically speed up Boolean satisfiability, a fundamental problem in computing with applications from chip design to cryptography.
New GPU Method Speeds Up Complex AI Optimization
A hybrid algorithm combines swarm intelligence and gradient methods to solve tough problems 10 to 100 times faster, with applications from physics to finance.
AI Drives Like Humans by Focusing on What Matters
A new AI method for autonomous vehicles mimics human attention, pruning 85% of visual data while maintaining 94% accuracy, making complex driving decisions feasible on standard hardware.
AI Adapts Chip Designs as Requirements Change
A new AI framework uses traceability links to update hardware code incrementally, reducing token usage by 23% and maintaining consistency without full regeneration.
Electric Circuits Reveal Hidden Quantum-Like Twists
Researchers have built a simple electric circuit that mimics exotic quantum behaviors, creating twisted energy patterns and new ways to control signals without complex materials.
AI Maps Indoor Spaces in Real Time for Navigation Assistance
A new method uses AI to create 3D maps with memory efficiency, enabling real-time tracking of objects and changes in dynamic environments for assistive navigation.
AI Achieves Consistent 3D Vision Across Multiple Cameras
A new method uses a cylindrical projection to align images from different angles, improving depth estimation for autonomous vehicles without needing extra sensors.
AI Tracker Adapts to Camera Glitches in Real Time
A new method uses a three-state switch to handle RGB, infrared, and invalid video feeds, preventing object loss during sensor failures and achieving 65 FPS performance.