Researchers have developed Auriga, a new task-agnostic agent harness designed to improve zero-shot robot manipulation by bridging the gap between planning and execution. Auriga integrates explicit task beliefs, precise 3D measurements, and dual-arm control, enabling models like GPT-6 Astra to achieve significantly higher success rates on benchmarks such as RoboDojo and RoboCasa365. The system monitors execution, stops the robot upon divergence, and allows for in-place recovery, leading to state-of-the-art performance. AI
IMPACT Enhances robot manipulation capabilities by improving the planning-execution loop for vision-language models.
RANK_REASON The cluster describes a technical report detailing a new research system for robot manipulation. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Daily Papers →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →