Related papers: LACUNA: Safe Agents as Recursive Program Holes

Task planning, the problem of sequencing actions to reach a goal from an initial state, is a core capability requirement for autonomous robotic systems. Whether large language models (LLMs) can serve as viable planners alongside classical…

LLM-powered agents are both a promising new technology and a source of complexity, where choices about models, tools, and prompting can affect their usefulness. While numerous benchmarks measure agent accuracy across domains, they mostly…

LLMs are increasingly used as general-purpose reasoners, but long inputs remain bottlenecked by a fixed context window. Recursive Language Models (RLMs) address this by externalising the prompt and recursively solving subproblems. Yet…

LLM agents with tool access can discover and exploit security vulnerabilities. This is known. What is not known is which features of a system prompt trigger this behaviour, and which do not. We present a systematic taxonomy based on…

Computer-use agents extend language models from text generation to persistent action over tools, files, and execution environments. Unlike chat systems, they maintain state across interactions and translate intermediate outputs into…

Agentic systems based on large language models (LLMs) operate not merely as text generators but as autonomous entities that dynamically retrieve information and invoke tools. This execution model shifts the attack surface from traditional…

Large Language Model (LLM)-based agentic systems have shown growing promise in tackling complex, multi-step tasks through autonomous planning, reasoning, and interaction with external environments. However, the stochastic nature of LLM…

Coding agents, which are LLM-driven agents specialized in software development, have become increasingly prevalent in modern programming environments. Unlike traditional AI coding assistants, which offer simple code completion and…

Large language model (LLM)-based agents have shown strong capabilities in using external tools to solve complex tasks. However, existing evaluations often overlook the temporal dimension of tool use, especially the impact of tool response…

AI agents powered by large language models (LLMs) are being deployed at scale, yet we lack a systematic understanding of how the choice of backbone LLM affects agent security. The non-deterministic sequential nature of AI agents complicates…

LLM agents do not act on raw interaction history; they act on a bounded decision state assembled by truncation, summarization, reordering, and rewriting. If directive-bearing state is dropped, weakened, or rebound during that step, an agent…

Large Language Model (LLM) based agents have demonstrated proficiency in multi-step interactions with graphical user interfaces (GUIs). While most research focuses on improving single-task performance, practical scenarios often involve…

Recent advances in language model (LM) agents and function calling have enabled autonomous, feedback-driven systems to solve problems across various digital domains. To better understand the unique limitations of LM agents, we introduce…

Large language Models (LLMs) have shown remarkable proficiency in code generation tasks across various programming languages. However, their outputs often contain subtle but critical vulnerabilities, posing significant risks when deployed…

The prevalent deployment of Large Language Model agents such as OpenClaw unlocks potential in real-world applications, while amplifying safety concerns. Among these concerns, the self-replication risk of LLM agents driven by objective…

Multi-agent systems coordinate LLM-based agents to perform tasks on users' behalf. In real-world applications, multi-agent systems will inevitably interact with untrusted inputs, such as malicious Web content, files, email attachments, and…

Most LLM safety work studies single-agent models, but many real applications rely on multiple interacting agents. In these systems, prompt segmentation and inter-agent routing create attack surfaces that single-agent evaluations miss. We…

LLM agents increasingly act on external systems, yet tool effects are immediate. Under failures, speculation, or contention, losing branches can leak unintended side effects with no safe rollback. We introduce Atomix, a runtime that…

Recent researches have shown that Large Language Models (LLMs) are susceptible to a security threat known as Backdoor Attack. The backdoored model will behave well in normal cases but exhibit malicious behaviours on inputs inserted with a…

AI agents that interact with the real world through tool calls pose fundamental safety challenges: agents might leak private information, cause unintended side effects, or be manipulated through prompt injection. To address these…