Tagged articles

RoboDojo

4 articles · Page 1 of 1

Sep 30, 2026 · Artificial Intelligence

RoboICL: Frozen GPT-6 Astra Achieves SOTA Robot Control via In-Context Learning

RoboICL enables frozen GPT-6 Astra to control dual-arm robots by providing demonstrations and the robot's own execution history as embodied context, achieving 50.64 average progress score on 30 RoboDojo tasks—surpassing prior methods—with detailed analysis of memory design, real-robot transfer, and 929 executable trajectories for inspection.

GPT-6 AstraRoboDojoRoboICL

0 likes · 19 min read

RoboICL: Frozen GPT-6 Astra Achieves SOTA Robot Control via In-Context Learning

Machine Heart

Sep 11, 2026 · Artificial Intelligence

EMERGE-Policy: Multi-Agent Framework Unifies VLA, World Models for Embodied AI

Tsinghua researchers propose EMERGE-Policy, a multi-agent framework integrating VLA, world models, and verifiers into a unified skill library with hierarchical agents and memory management, achieving state-of-the-art results on LIBERO and RoboDojo benchmarks and robust real-world cup-stacking under disturbances.

EMERGE-PolicyLIBERO benchmarkRoboDojo

0 likes · 12 min read

EMERGE-Policy: Multi-Agent Framework Unifies VLA, World Models for Embodied AI

Machine Heart

Sep 6, 2026 · Artificial Intelligence

VLAct: 16 GPUs, 20% Data Beats GR00T N1.6 in Cross-Embodiment Transfer

VLAct introduces a representation-centric continued pre-training framework for Vision-Language-Action models, achieving 92.5% on RoboTwin 2.0 and surpassing all World Action Models on RoboDojo using only 16 GPUs and open data; with 20% downstream data it outperforms GR00T N1.6 on unseen GR-1 robot.

RoboDojoRoboTwinVLA

0 likes · 10 min read

VLAct: 16 GPUs, 20% Data Beats GR00T N1.6 in Cross-Embodiment Transfer

Machine Heart

Jul 17, 2026 · Artificial Intelligence

Can General-Purpose Robots Arrive Soon? Inside the RoboDojo “Embodied Everest” Benchmark

The RoboDojo benchmark evaluates 30 robot manipulation policies across 42 simulated and 18 real-world tasks, revealing that the best models achieve only 8.8% success in simulation and 12.8% in reality, far behind human experts, and highlights gaps in generalization, memory, precision, long‑horizon execution, and open‑semantic understanding.

RoboDojoreal-world evaluationrobotic manipulation

0 likes · 9 min read

Can General-Purpose Robots Arrive Soon? Inside the RoboDojo “Embodied Everest” Benchmark