Top pick  ·  LLMs & Language Models

World In-Context Learning for Agentic Robot Manipulation

Robots learn manipulation from a single demo by rehearsing in simulation, writing feedback to context, achieving 83.4% success without gradient updates.

Why it matters: Allows rapid skill acquisition for robots without costly retraining, leveraging simulation to adapt to new scenes.

Read the breakdown →  Code

Trending  ·  Other ML

Learning Heterogeneous Tactile Representation for General Dexterous Manipulation

LingBot-Touch learns a unified tactile representation that works across different sensor types and improves dexterous manipulation.

Why it matters: Developers can reuse one tactile model for many sensors, simplifying system integration and boosting robot skill performance.

Read the breakdown →