Drone Technology

Technology

NVIDIA Cosmos 3 Post-Training Completed in One Day: AI Coding Agents Boost Visual Reasoning Model Accuracy Beyond 90%

Autonomous coding AI agents are transforming visual reasoning model development by automating data formatting, container setup, training scripts, and hyperparameter tuning — tasks that traditionally take days. Applied to NVIDIA Cosmos 3, the entire post-training pipeline can now be completed in a single day with minimal human intervention, pushing model accuracy above 90%. The breakthrough holds particular significance for drone and autonomous vehicle applications that rely on high-precision computer vision.

AI自動駕駛電腦視覺

about 2 months ago

Source: NVIDIA Robotics Blog

Technology

Technology

Cellebrite and SkySafe Partner to Integrate Counter-Drone Systems with Digital Forensics

Orbion Technologies Joins DronePort Rotterdam as Innovation Partner to Advance Drone DAA Capabilities

High Lander and Thirdeye Systems Partner to Develop Ground-Based DAA Services for Scalable BVLOS Operations

Swedish Startup Unveils 'Radio-Silent' Counter-Drone Interceptor System

Farnborough Airshow: Eve Air Mobility Showcases Full-Scale Mock-Up and Flight Test Progress

AT&T and Ericsson Demo 5G Drone Detection Technology Outside AT&T Stadium

MQ-25A Stingray Unmanned Tanker Completes Second Test Flight

LiDAR Scanning Technology Powers Digital Models Behind Vivid Sydney's Spectacular Light Shows

AI Agents Build Virtual Training Grounds to Generate Critical Robotics Training Data

Historic First: U.S. Corsair Autonomous Surface Vessels Strike Iranian Naval Base at Bandar Abbas

NVIDIA Cosmos 3 Post-Training Completed in One Day: AI Coding Agents Boost Visual Reasoning Model Accuracy Beyond 90%

This Inflatable Fixed-Wing Drone Looks Just Like Baymax — and Flies for Over 10 Hours