arXiv subjects

Linxuan He

Publications and source records attributed to Linxuan He.

3 recordsLinked to original sources

arXiv2026

Self-Evolving Learning for Embodied AI with Criticality Model

Despite rapid advances in policy pretraining, embodied AI systems routinely plateau during task-specific finetuning. The root cause lies in how finetuning data are collected: the default pipeline gathers data randomly, treating every sample as informative. Datasets become dominated by nominal scenarios, while rare failure cases--the most valuable for improvement--are missed. We propose a self-evolving method that breaks this plateau. Our core insight is that a state-wise criticality model, learned from the policy's own execution outcomes to predict the probability of future failure, can guide importance sampling toward failure-prone scenarios. After replacing redundant nominal scenarios with diverse failure-prone ones, importance weights are used to resample the data during training. This effectively preserves an unbiased learning objective while fundamentally increasing the information density of the training pool. Across quadrupedal locomotion, multi-task manipulation, vision-language-action benchmarks, and a real-robot task, our method reduces failure rates by 51--67% relative to trained baselines and by 8-25% relative to state-of-the-art vision-language-action models.

eess.SY↗

arXiv2025

Efficient Safety Verification of Autonomous Vehicles with Neural Network Operator

When autonomous vehicles encounter untrained scenarios, ensuring safety hinges on effective safety verification to prevent accidents stemming from unexpected model decisions. Reachability analysis, a method of safety verification, offers relatively high precision but at the cost of significant computational complexity. Our method leverages end-to-end neural network operators to compute reachable sets, replacing traditional mathematical set operators, thereby achieving higher efficiency in safety verification without substantially compromising accuracy or increasing conservativeness. We define vehicle dynamics on discrete time series and detail the safety verification process and safety standard based on reachable sets. Experimental evaluations conducted in several typical road driving scenarios demonstrate the superior efficiency performance of our proposed operator over classical methods.

eess.SY↗

arXiv2025

Towards provable probabilistic safety for scalable embodied AI systems

Embodied AI systems, comprising AI models and physical plants, are increasingly prevalent across various applications. Due to the rarity of system failures, ensuring their safety in complex operating environments remains a major challenge, which severely hinders their large-scale deployment in safety-critical domains, such as autonomous vehicles, medical devices, and robotics. While achieving provable deterministic safety-verifying system safety across all possible scenarios-remains theoretically ideal, the rarity and complexity of corner cases make this approach impractical for scalable embodied AI systems. Instead, empirical safety evaluation is employed as an alternative, but the absence of provable guarantees imposes significant limitations. To address these issues, we argue for a paradigm shift to provable probabilistic safety that integrates provable guarantees with progressive achievement toward a probabilistic safety boundary on overall system performance. The new paradigm better leverages statistical methods to enhance feasibility and scalability, and a well-defined probabilistic safety boundary enables embodied AI systems to be deployed at scale. In this Perspective, we outline a roadmap for provable probabilistic safety, along with corresponding challenges and potential solutions. By bridging the gap between theoretical safety assurance and practical deployment, this Perspective offers a pathway toward safer, large-scale adoption of embodied AI systems in safety-critical applications.

eess.SY↗