Cong Wang

The paper introduces ARES, a novel and practical gradient inversion attack that reconstructs sensitive training samples from large batch updates in Federated Learning without requiring architectural modifications.

This survey provides a comprehensive, structured review of safety research in Embodied AI, analyzing attacks and defenses across the entire embodied pipeline to guide the development of safe, robust, and reliable real-world agents.

STARE introduces a novel hierarchical reinforcement learning framework that treats the entire image generation process (denoising trajectory) as an attack surface, significantly improving the detection of multi-modal toxicity vulnerabilities in Vision-Language Models.

Sandlock is a lightweight, unprivileged Linux process sandbox that enforces fine-grained policies over filesystem, network, and syscalls for running untrusted AI agent code, achieving strong isolation without requiring root privileges or complex virtualization.

FedMPT introduces a novel federated learning framework for Multi-Label Recognition (MLR) using Vision-Language Models (VLMs) by leveraging generalizable conditions to mitigate label overfitting and improve robustness.

This paper analyzes the x402 payment protocol, revealing critical synchronization and security flaws that allow attackers to exploit payment systems and force merchants to subsidize compute costs.

This paper analyzes the x402 payment protocol, revealing systemic vulnerabilities in state synchronization and signature design that allow attackers to exploit payment systems for resource leakage in the AI economy.

TeeDAO introduces a novel three-layer framework that autonomously organizes and manages multiple heterogeneous Trusted Execution Environments (TEEs) to provide robust, distributed-trust systems with high throughput and strong security guarantees.

This paper proposes a new method for agentic Reinforcement Learning called Agentic Procedural Policy Optimization (APPO) that improves tool-use capabilities by assigning credit to fine-grained decision points.