Related papers: Do Coding Agents Understand Least-Privilege Author…
We formalize trust calibration for agentic tool use (deciding when an automated agent's proposed action may execute autonomously versus require human approval) as a preference-learning problem. A policy gateway maintains a Gaussian-process…
Repository-level issue resolution benchmarks have become a standard testbed for evaluating LLM-based agents, yet success is still predominantly measured by test pass rates. In practice, however, acceptable patches must also comply with…
Modern operating systems provide powerful mandatory access control mechanisms, yet they largely reason about who executes code rather than how execution originates. As a result, processes launched remotely, locally, or by background…
Understanding code represents a core ability needed for automating software development tasks. While foundation models like LLMs show impressive results across many software engineering challenges, the extent of their true semantic…
Many scenarios where agents with restrictions compete for resources can be cast as maximum matching problems on bipartite graphs. Our focus is on resource allocation problems where agents may have restrictions that make them incompatible…
Attribute-based methods, such as attribute-based access control and attribute-based encryption, make decisions based on attributes possessed by a subject rather than the subject's identity. While this allows for anonymous authorization --…
We consider a scenario in which an autonomous agent carries out a mission in a stochastic environment while passively observed by an adversary. For the agent, minimizing the information leaked to the adversary regarding its high-level…
AI agents dynamically acquire tools, orchestrate sub-agents, and transact across organizational boundaries, yet no existing security layer verifies what an agent can do, whether it executed what it claims, or what happened in a multi-agent…
Terminal agents are increasingly capable of executing complex, long-horizon tasks autonomously from a single user prompt. To do so, they must interpret instructions encountered in the environment (e.g., README files, code comments, stack…
Reusable skills are becoming a common interface for extending large language model agents, packaging procedural guidance with access to files, tools, memory, and execution environments. However, this modularity introduces attack surfaces…
Hierarchical predictive processing explains adaptive behaviour through precision-weighted inference. Explicit belief revision often fails to produce corresponding changes in stress reactivity or autonomic regulation. This asymmetry suggests…
Adapting pre-trained deep learning models to customized tasks has become a popular choice for developers to cope with limited computational resources and data volume. More specifically, probing--training a downstream head on a pre-trained…
Agentic systems increasingly rely on language models to monitor their own behavior. For example, coding agents may self critique generated code for pull request approval or assess the safety of tool-use actions. We show that this design…
AI agents may soon become capable of autonomously completing valuable, long-horizon tasks in diverse domains. Current benchmarks either do not measure real-world tasks, or are not sufficiently difficult to meaningfully measure frontier…
LLM agents process trusted instructions, retrieved records, and tool observations through a common generative channel. This conflates data flow with authority: an untrusted string can affect a secret-bearing response or an action proposal…
Large language models increasingly operate as autonomous agents that select and invoke tools from large registries. We identify a critical gap: when unauthorized tools are visible in an agent's context, models select them in adversarial…
Although Large Language Models (LLMs) have evolved from text generators into the cognitive core of modern AI systems, their inherent lack of authorization awareness exposes these systems to catastrophic risks, ranging from unintentional…
Deep reinforcement learning has emerged as a powerful tool for obtaining high-performance policies. However, the safety of these policies has been a long-standing issue. One promising paradigm to guarantee safety is a shield, which shields…
Adversarial training has been widely studied in recent years due to its role in improving model robustness against adversarial attacks. This paper focuses on comparing different distributed adversarial training algorithms--including…
Large language models (LLMs) are increasingly deployed as tool-using agents, shifting safety concerns from harmful text generation to harmful task completion. Deployed systems often condition on user profiles or persistent memory, yet agent…