UT Austin open-sources OopsieVerse, a damage-aware safety benchmark for household robots
DynamicWebPaige · x · 2026-11-10
UT Austin released OopsieVerse (RSS 2026), an open-source benchmark tackling robots that finish tasks while breaking things.
- DamageSim: a simulator-agnostic plugin (MuJoCo/RoboCasa, BEHAVIOR-1K) converting contact forces, heat and liquid exposure into measurable mechanical, thermal and fluid damage scores.
- OopsieBench: 32 household tasks contrasting easy-but-risky vs safer strategies.
- Supports safer data collection, damage-aware imitation/RL, VLA safety evals and sim-to-real transfer.
More from Embodied
- Perceptron open-sources EgoHome, a sub-second egocentric household-video benchmark — lukas_m_ziegler · 2026-11-11
- Anthropic touts Fable 5.2 robot control gains on RoboDojo; real-world box folding disappoints — _Stocko_ · 2026-11-11
- AI2 open-sources ToyBlocks to benchmark robotic world models by task-relevant fidelity — allen_ai · 2026-11-11
- OpenTac: open-source ecosystem turns any robot surface into a tactile skin — _krishna_murthy · 2026-11-11
- NVIDIA launches open Alpamayo 2 Edge for real-time in-vehicle AI reasoning — nvidia · 2026-11-11
- NVIDIA open-sources GR00T 2 robot foundation model with 100K hours of training data — DrJimFan · 2026-11-11