analiza aiqbench
The paper is a research agenda, not a result. It proposes a "Foundation Model Assurance Stack" that pairs neuro-symbolic architectures, conformance-based testing, and compositional verification with runtime monitoring, so that foundation models in autonomous vehicles and UAVs are checked wherever correctness can be formally specified. Its main contribution is framing and open challenges (specification engineering, abstraction fidelity, evidence composition, lifecycle certification) rather than validated methods. For AI readers, the practical question is whether symbolic boundaries can give certifiable guarantees around stochastic learned components, which remains unproven.
Assuring Learning-Enabled Cyber-Physical Systems with Foundation Models: A Neuro-Symbolic Agenda for Testing, Verification, and Runtime Assurance
arXiv:2610.39547v1 Announce Type: new Abstract: Learning-enabled cyber-physical systems (CPS), including autonomous vehicles and UAVs, increasingly incorporate foundation models, yet fragmented assurance practices provide insufficient support for system-level guarantees. We argue that restoring engineering control across stochastic neural, symbolic, and physical boundaries requires a shift from model-centric evaluation to system-level assurance. We introduce the Foundation Model Assurance Stack, connecting three directions: (i) neuro-symbolic architectures that constrain foundation models through explicit symbolic world models, reasoning layers, and interfaces; (ii) semantically grounded, conformance-based testing that combines formal specifications and symbolic world models with foundation-model-assisted specification mining, scenario generation, and test synthesis; and (iii) compositional verification and runtime assurance that target formally tractable components and interfaces through modular contracts, equivalence checking, verified synthesis, uncertainty-aware reasoning, and runtime monitoring and enforcement. Symbolic abstractions establish assurance boundaries between stochastic learning and formal reasoning, connecting testing, verification, and runtime evidence across the system lifecycle. Foundation models serve as constrained assistants whose outputs are independently checked wherever correctness can be formally specified. We identify research challenges in specification engineering, abstraction fidelity, evidence composition, and lifecycle certification, and outline a staged agenda toward trustworthy, deployable, and certifiable learning-enabled CPS.
Authors:Xi Zheng, Ivan Ruchkin, Hong Jin Kang, Tevfik Bultan, Vijay Ganesh, Taylor T. Johnson, Ziyang Li, Lei Ma, Aloysius K. Mok, Daniel Neider, Ruzica Piskac, Nalini Venkatasubramanian, Simon Thompson, Sanjoy Baruah
Abstract:Learning-enabled cyber-physical systems (CPS), including autonomous vehicles and UAVs, increasingly incorporate foundation models, yet fragmented assurance practices provide insufficient support for system-level guarantees. We argue that restoring engineering control across stochastic neural, symbolic, and physical boundaries requires a shift from model-centric evaluation to system-level assurance. We introduce the Foundation Model Assurance Stack, connecting three directions: (i) neuro-symbolic architectures that constrain foundation models through explicit symbolic world models, reasoning layers, and interfaces; (ii) semantically grounded, conformance-based testing that combines formal specifications and symbolic world models with foundation-model-assisted specification mining, scenario generation, and test synthesis; and (iii) compositional verification and runtime assurance that target formally tractable components and interfaces through modular contracts, equivalence checking, verified synthesis, uncertainty-aware reasoning, and runtime monitoring and enforcement. Symbolic abstractions establish assurance boundaries between stochastic learning and formal reasoning, connecting testing, verification, and runtime evidence across the system lifecycle. Foundation models serve as constrained assistants whose outputs are independently checked wherever correctness can be formally specified. We identify research challenges in specification engineering, abstraction fidelity, evidence composition, and lifecycle certification, and outline a staged agenda toward trustworthy, deployable, and certifiable learning-enabled CPS.
Robotics (cs.RO); Systems and Control (eess.SY)
arXiv:2610.39547 [cs.RO]
(or arXiv:2610.39547v1 [cs.RO] for this version)
Text integral sub licența deschisă a publicației · 230 cuvinte