Two New Robotics VLA Studies
shahdhruv_ · x · 2026-07-12
This post covers the 2nd and 3rd robotics papers from RSS:
- Long-Context Robot Imitation Learning by Focusing on Key History Frames: Discusses the challenges of training history-conditioned policies, proposing that models focus on key historical frames to handle complex tasks requiring long-horizon understanding, like treasure hunts.
- LAP: Language-Action Pre-training Enables Zero-Shot Cross-Embodiment Transfer: Proposes using language actions and an aligned action space to achieve zero-shot transfer across embodiments, claiming a new SOTA for open-weight VLA in academia.
More from Embodied
- NVIDIA open-sources FLORA, a robotics data engine ~125x cheaper than GPT-6 Astra — nvidia · 2026-10-27
- Embodied AI startups need a different operational discipline, investor says — ___Mufasaa · 2026-10-27
- π0.5 study: raising demos from 10 to 438 lifts Place Ball in Bowl success from 7% to 83% — stingning · 2026-10-27
- GPT-6 Astra controls dexterous hands zero-shot at just 5.2% success in 500 trials — stingning · 2026-10-27
- HandsOn benchmark: 66 verified tasks and 18,000 human demos for dexterous manipulation — stingning · 2026-10-27
- DexPort: a handheld interface with a real robot hand for dexterous-hand data collection — pabbeel · 2026-10-27