arXiv subjects

Silei Ren

Publications and source records attributed to Silei Ren.

3 recordsLinked to original sources

arXiv2026

Towards Trustworthy Physical AI: From Theory to Practice Across Life Cycle

Physical AI refers to AI systems that understand, reason about, and act in accordance with the physical world and its underlying laws, dynamics, and constraints. Unlike conventional AI systems, physical AI interacts continuously with uncertain physical environments, and its actions produce consequences that are physically irreversible. As existing trustworthy AI frameworks have been developed primarily for digital AI systems, they do not fully capture the distinctive challenges of physical AI, such as physical safety, cyber-physical security, and physical manufacturing process. To address this gap, we present a survey of trustworthy physical AI principles. First, we characterize the core capabilities and challenges of physical AI. Second, we examine the role of physics in AI. Third, we trace the end-to-end physical AI life cycle across five core stages and introduce Trustworthy Physical AI Operationalization (T-PAIO). Fourth, we develop the Trustworthy Physical AI (T-PAI) framework, a theoretical framework that organizes key trustworthiness principles and provides a foundation for governing trustworthy physical AI systems.

cs.AI↗

arXiv2025

An Algebraic Approach to Asymmetric Delegation and Polymorphic Label Inference (Technical Report)

Language-based information flow control (IFC) enables reasoning about and enforcing security policies in decentralized applications. While information flow properties are relatively extensional and compositional, designing expressive systems that enforce such properties remains challenging. In particular, it can be difficult to use IFC labels to model certain security assumptions, such as semi-honest agents. Motivated by these modeling limitations, we study the algebraic semantics of lattice-based IFC label models, and propose a semantic framework that allows formalizing asymmetric delegation, which is partial delegation of confidentiality or integrity. Our framework supports downgrading of information and ensures their safety through nonmalleable information flow (NMIF). To demonstrate the practicality of our framework, we design and implement a novel algorithm that statically checks NMIF and a label inference procedure that efficiently supports bounded label polymorphism, allowing users to write code generic with respect to labels.

cs.PL↗

arXiv2020

The Price of Tailoring the Index to Your Data: Poisoning Attacks on Learned Index Structures

The concept of learned index structures relies on the idea that the input-output functionality of a database index can be viewed as a prediction task and, thus, be implemented using a machine learning model instead of traditional algorithmic techniques. This novel angle for a decades-old problem has inspired numerous exciting results in the intersection of machine learning and data structures. However, the main advantage of learned index structures, i.e., the ability to adjust to the data at hand via the underlying ML-model, can become a disadvantage from a security perspective as it could be exploited. In this work, we present the first study of poisoning attacks on learned index structures. The required poisoning approach is different from all previous works since the model under attack is trained on a cumulative distribution function (CDF) and, thus, every injection on the training set has a cascading impact on multiple data values. We formulate the first poisoning attacks on linear regression models trained on the CDF, which is a basic building block of the proposed learned index structures. We generalize our poisoning techniques to attack a more advanced two-stage design of learned index structures called recursive model index (RMI), which has been shown to outperform traditional B-Trees. We evaluate our attacks on real-world and synthetic datasets under a wide variety of parameterizations of the model and show that the error of the RMI increases up to $300\times$ and the error of its second-stage models increases up to $3000\times$.

cs.CR↗