相关论文: WebInject: Prompt Injection Attack to Web Agents

LLM workflows, which coordinate structured calls to individual LLMs/agents to achieve a particular goal, offer a promising path towards building powerful AI systems that can tackle diverse tasks. However, existing approaches for building…

Multi-modal large language models (MLLMs) have emerged as powerful tools for analyzing Internet-scale image data, offering significant benefits but also raising critical safety and societal concerns. In particular, open-weight MLLMs may be…

Modern threats have emerged from the prevalence of social networks. Hostile actors, such as extremist groups or foreign governments, utilize these networks to run propaganda campaigns with different aims. For extremists, these campaigns are…

LLM agents now draw on growing skill libraries to handle complex tasks. However, injecting more skills does not always improve task completion and can even degrade it. Existing methods still treat skill injection as a static step, selecting…

Vision-language models (VLMs) (e.g. CLIP, LLaVA) are trained on large-scale, lightly curated web datasets, leading them to learn unintended correlations between semantic concepts and unrelated visual signals. These associations degrade…

Indirect Prompt Injection attacks exploit the inherent limitation of Large Language Models (LLMs) to distinguish between instructions and data in their inputs. Despite numerous defense proposals, the systematic evaluation against adaptive…

Large language models (LLMs) have transformed the development of embodied intelligence. By providing a few contextual demonstrations, developers can utilize the extensive internal knowledge of LLMs to effortlessly translate complex tasks…

The web is littered with images, once created for human consumption and now increasingly interpreted by agents using vision-language models (VLMs). These agents make visual decisions at scale, deciding what to click, recommend, or buy. Yet,…

As AI agents powered by Large Language Models (LLMs) become increasingly versatile and capable of addressing a broad spectrum of tasks, ensuring their security has become a critical challenge. Among the most pressing threats are prompt…

Generative large vision-language models (LVLMs) have recently achieved impressive performance gains, and their user base is growing rapidly. However, the security of LVLMs, in particular in a long-context multi-turn setting, is largely…

LLM agents are evolving rapidly, powered by code execution, tools, and the recently introduced agent skills feature. Skills allow users to extend LLM applications with specialized third-party code, knowledge, and instructions. Although this…

We demonstrate how AI-powered cybersecurity tools can be turned against themselves through prompt injection attacks. Prompt injection is reminiscent of cross-site scripting (XSS): malicious text is hidden within seemingly trusted content,…

Recently, deep networks have achieved impressive semantic segmentation performance, in particular thanks to their use of larger contextual information. In this paper, we show that the resulting networks are sensitive not only to global…

In-context learning (ICL) has emerged as a powerful paradigm leveraging LLMs for specific downstream tasks by utilizing labeled examples as demonstrations (demos) in the preconditioned prompts. Despite its promising performance, crafted…

Memory makes LLM-based web agents personalized, powerful, yet exploitable. By storing past interactions to personalize future tasks, agents inadvertently create a persistent attack surface that spans websites and sessions. While existing…

Agents powered by large language models (LLMs) have demonstrated strong capabilities in a wide range of complex, real-world applications. However, LLM agents with a compromised memory bank may easily produce harmful outputs when the past…

Agent skills are increasingly used to extend LLM agents with task-specific instructions, executable scripts, and auxiliary resources. While improving reusability, this modular design also introduces a new supply-chain attack surface: a…

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…

Recent advances in operating system (OS) agents have enabled vision-language models (VLMs) to directly control a user's computer. Unlike conventional VLMs that passively output text, OS agents autonomously perform computer-based tasks in…

Large language models have become increasingly prominent, also signaling a shift towards multimodality as the next frontier in artificial intelligence, where their embeddings are harnessed as prompts to generate textual content.…