10 papers
SafeVLA-Bench: A Benchmark for the Success-Safety Gap in Vision-Language-Action Models
Jialiang Fan, Weizhe Xu, Oleg Sokolsky +2
Vision-language-action (VLA) benchmarks measure whether a policy completes a requested manipulation task, but binary success can hide safety-relevant trajectory behavior: reaching…
SafeGen-LLM: Enhancing Safety Generalization in Task Planning for Robotic Systems
Jialiang Fan, Weizhe Xu, Mengyu Liu +3
Safety-critical task planning in robotic systems remains challenging: classical planners suffer from poor scalability, Reinforcement Learning (RL)-based methods generalize poorly,…
Conservative Perception Models for Probabilistic Verification
Matthew Cleaveland, Pengyuan Lu, Oleg Sokolsky +2
Verifying the behaviors of autonomous systems with learned perception components is a challenging problem due to the complexity of the perception and the uncertainty of operating e…
Safety Monitoring for Learning-Enabled Cyber-Physical Systems in Out-of-Distribution Scenarios
Vivian Lin, Ramneet Kaur, Yahan Yang +6
The safety of learning-enabled cyber-physical systems is compromised by the well-known vulnerabilities of deep neural networks to out-of-distribution (OOD) inputs. Existing literat…
Distributionally Robust Statistical Verification with Imprecise Neural Networks
Souradeep Dutta, Michele Caprio, Vivian Lin +5
A particularly challenging problem in AI safety is providing guarantees on the behavior of high-dimensional autonomous systems. Verification approaches centered around reachability…
AR-Pro: Counterfactual Explanations for Anomaly Repair with Formal Properties
Xiayan Ji, Anton Xue, Eric Wong +2
Anomaly detection is widely used for identifying critical errors and suspicious behaviors, but current methods lack interpretability. We leverage common properties of existing meth…