相关论文: Session Risk Memory (SRM): Temporal Authorization …

How much does a trained RL policy actually use its past observations? We propose \emph{Temporal Range}, a model-agnostic metric that treats first-order sensitivities of multiple vector outputs across a temporal window to the input sequence…

Although LLM-based agents, powered by Large Language Models (LLMs), can use external tools and memory mechanisms to solve complex real-world tasks, they may also introduce critical security vulnerabilities. However, the existing literature…

LLMs are increasingly equipped with safety alignment mechanisms, yet recent studies demonstrate that they remain vulnerable to jailbreaking attacks that elicit harmful behaviors without explicit policy violations. While a growing body of…

Large Language Models (LLMs) have transformed task automation and content generation across various domains while incorporating safety filters to prevent misuse. We introduce a novel jailbreaking framework that employs distributed prompt…

Visual language models (VLMs) have made significant progress in image captioning tasks, yet recent studies have found they are vulnerable to backdoor attacks. Attackers can inject undetectable perturbations into the data during inference,…

The transition of Large Language Models (LLMs) from passive code generators to autonomous agents introduces significant safety risks, specifically regarding destructive commands and inconsistent system states. Existing commercial solutions…

Autonomous driving policy learning with reinforcement learning (RL) is fundamentally limited by low sample efficiency, weak generalization, and a dependence on unsafe online trial-and-error interactions. Although safe RL introduces explicit…

Ensuring the safety of embodied AI agents during task planning is critical for real-world deployment, especially in household environments where dangerous instructions pose significant risks. Existing methods often suffer from either high…

Understanding risk in autonomous driving requires not only perception and prediction, but also high-level reasoning about agent behavior and context. Current Vision Language Model (VLM)-based methods primarily ground agents in static images…

In this paper, we consider the problem of safety assessment for Markov decision processes without explicit knowledge of the model. We aim to learn probabilistic safety specifications associated with a given policy without compromising the…

Recent advances in foundation models have transformed LLMs from passive conversational systems into autonomous agents capable of reasoning and tool execution. While these capabilities unlock substantial practical value, they also introduce…

Large language model (LLM) agents achieve impressive single-task performance but commonly exhibit repeated failures, inefficient exploration, and limited cross-task adaptability. Existing reflective strategies (e.g., Reflexion, ReAct)…

Current LLM safety defenses fail under decomposition attacks, where a malicious goal is decomposed into benign subtasks that circumvent refusals. The challenge lies in the existing shallow safety alignment techniques: they only detect harm…

Advanced Driver Assistance Systems (ADAS) increasingly rely on learning-based perception, yet safety-relevant failures often arise without component malfunction, driven instead by partial observability and semantic ambiguity in how risk is…

Simulation-based testing is crucial for validating autonomous vehicles (AVs), yet existing scenario generation methods either overfit to common driving patterns or operate in an offline, non-interactive manner that fails to expose rare,…

Modern LLM agents combine long-term memory for personalization with tool-calling interfaces for taking actions in the world -- a combination underpinning contemporary production systems. We study a previously unexamined failure of this…

Recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) have demonstrated significant potential in single-turn reasoning tasks. With the paradigm shift toward self-evolving agentic learning, models are increasingly expected…

This paper presents a novel online framework for safe crowd-robot interaction based on risk-sensitive stochastic optimal control, wherein the risk is modeled by the entropic risk measure. The sampling-based model predictive control relies…

Non-Markovian Reinforcement Learning (RL) tasks present significant challenges, as agents must reason over entire trajectories of state-action pairs to make optimal decisions. A common strategy to address this is through symbolic…

Memory systems enable otherwise-stateless LLM agents to persist user information across sessions, but also introduce a new attack surface. We characterize the Trojan Hippo attack, a class of persistent memory attacks that operates in a more…