Robot learning paper club #5: π0.5 fine-tuned on a real manufacturing task
DominiqueCAPaul · x · 2026-09-25
Recap of the 5th robot learning paper club, six talks covering co-design, navigation, and manipulation:
- Transformer Transformer: a unified model for motion-conditioned robot co-design
- Hierarchical Open-Vocabulary 3D Scene Graphs for language-grounded robot navigation
- GaP: Graph-as-Policy, a multi-agent self-learning harness for variational automation tasks
- Xing Li (FlexionAI) presented his thesis paper From a Single Demonstration to a General Policy for Contact-Rich Manipulation
- Disney Research's Michael Baumgartner explained teaching robots to dance with CoCo-InEKF: state estimation with learned contact covariances in dynamic, contact-rich scenarios
- Aurel Arnold shared new empirical results fine-tuning π0.5 on a real manufacturing task at Dream Machines
The next session is upcoming.
More from Embodied
- BracketBot robot autonomously navigates via video-built maps to scan shelves — broodsugar · 2026-09-25
- Smart light fixtures as an always-powered AI sensor platform for the home — K-enthusiast24 · 2026-09-25
- Meta is putting its Muse agent into AI glasses, aiming to make the interface disappear — TansuYegen · 2026-09-25
- Agents turn stock ESP32 kits into anything: why familiar forms guide what to build — genmon · 2026-09-25
- FANUC cobot takes plain-language commands, trained in Isaac Sim before hardware existed — lukas_m_ziegler · 2026-09-25
- The Gaussian is enough: Toyota study finds non-Gaussian priors don't help fine-tuning LBMs — _krishna_murthy · 2026-09-25