Robotics Posts

R&B-EnCoRe: Self-Improving Pretraining of Embodied Reasoning Vision-Language-Action Models

R&B-EnCoRe is a self-improvement pretraining cycle for embodied reasoning VLAs. Instead of relying on a fixed reasoning template, the model generates its own candidate reasoning, refines it by measuring how much each piece helps predict the correct action, and bootstraps a stronger policy.

How Machine Learning Brings
Unconventional Robots to Life

Xiaomeng Xu

Data-driven methods make it easy to explore unconventional robot design to unlock novel capabilities. Meet RoboPanoptes: The All-Seeing Robot with Whole-body Dexterity, and Dynamics-Guided Diffusion Model for Sensor-less Robot Manipulator Design!

Productive Struggle: The Future of Human Learning in the Age of AI

Rose E. Wang and Megha Srivastava

What happens to human learning when superhuman intelligence is as accessible as a Google search?

MiniVLA: A Better VLA with a Smaller Footprint

Suneel Belkhale and Dorsa Sadigh

Reducing OpenVLA's parameters 7x, and improving the input and output representation space.

Stanford AI Lab Papers and Talks at RSS 2023

Compiled by Drew A. Hudson

All the great work from the Stanford AI Lab accepted at RSS 2023, all in one place.

Self-Improving Robots: Embracing Autonomy in Robot Learning

Archit Sharma

Learning methods for robotics have a data problem. Autonomous robotic systems can deliver orders of magnitude more real-world embodied data than we have.

Stanford AI Lab Papers and Talks at CVPR 2023

Compiled by Drew A. Hudson

All the great work from the Stanford AI Lab accepted at CVPR 2023, all in one place.

Learning to Imitate

Divyansh Garg

SOTA methods for Imitation Learning are heavily data-inefficient or rely on unstable Adversarial training. We present a novel framework that introduces simple, stable, and data-efficient learning from few experts that scales well to complex environments.

Reward Isn't Free: Supervising Robot Learning with Language and Video from the Web

Suraj Nair

Where do the rewards for robotic reinforcement learning come from? In this blog post we study how using crowdsourced language annotations and videos of humans, we can learn reward functions in a scalable way and enable them to generalize more broadly.

What Matters in Learning from Offline Human Demonstrations for Robot Manipulation

Ajay Mandlekar

We conducted an extensive study of six offline learning algorithms for robot manipulation on five simulated and three real-world multi-stage manipulation tasks of varying complexity, and with datasets of varying quality. Our study analyzes the most critical challenges when learning from offline h...