News

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

21 days ago

Source: NVIDIA Robotics Blog

All Articles

Trump Claims Investigation Into Iranian Shahed Drone Storage in Cuba: What It Could Mean

Queensland Surveyors Board Issues Compliance Reminder to Mine Surveyors

OpenAI Researcher Miles Wang in Talks to Launch AI Drug Discovery Startup at $2 Billion Valuation

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

From Drones to Psychological Warfare: Kyiv Hosts Cognitive Influence & Resilience Forum

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

Mapping Australia's Eastern Seabed: CSIRO Research Vessel Launches Key Coastal Survey Mission

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

Leaderboard Lessons: What 5,000 Kaggle Competitors Taught Us About Improving AI Reasoning

New Balance Niobium Concept 1: A 3-in-1 Hiking Boot That Unzips Into a Camp Slipper