UC Berkeley's BAIR published its CoRL 2026 lineup: 33 main-conference papers, 4 late-breaking posters and 10 orals/spotlights, with Professor Negar Mehr delivering an Early Career Keynote. The post signals Berkeley's continued output density in robot learning, but is a schedule roundup without method or result details.
Robot Learning
机器人学习
Research field applying machine learning to robot perception, planning and control.
Curated Intelligence
Ai2 has released ToyBlocks, a controlled simulation environment for robot learning and evaluation, with code, datasets, models, and documentation released together. The goal is to let researchers systematically study what a robot has actually learned and why it fails, moving robot learning beyond narrow demonstrations toward transferable capability.
Hugging Face released LeRobot v0.6.0 to close the robot learning loop via an 'Imagine, Evaluate, Improve' framework. The update introduces policies that predict future states, reward models for success detection, and a deployment CLI that converts failures into training data, alongside six new simulation benchmarks.
The Hugging Face LeRobot project released LeRobotDataset v3.0, designed to optimize support for large-scale robotics datasets. This update focuses on improving data loading and storage efficiency to accommodate more complex embodied AI training scenarios.
The article title suggests using video encoding to scale robotics datasets, but the body text is missing. Specific methods, data scales, and performance gains cannot be evaluated.