Related papers: Continuous Discovery of Vulnerabilities in LLM Ser…

Large Language Model (LLM) serving systems remain fundamentally fragile, where frequent hardware faults in hyperscale clusters trigger disproportionate service outages in the software stack. Current recovery mechanisms are prohibitively…

Our team, All You Need Is A Fuzzing Brain, was one of seven finalists in DARPA's Artificial Intelligence Cyber Challenge (AIxCC), placing fourth in the final round. During the competition, we developed a Cyber Reasoning System (CRS) that…

In the context of Machine Learning as a Service (MLaaS) clouds, the extensive use of Large Language Models (LLMs) often requires efficient management of significant query loads. When providing real-time inference services, several…

When users query proprietary LLM APIs, they receive outputs with no cryptographic assurance that the claimed model was actually used. Service providers could substitute cheaper models, apply aggressive quantization, or return cached…

Recent work has shown that gradient updates in federated learning (FL) can unintentionally reveal sensitive information about a client's local data. This risk becomes significantly greater when a malicious server manipulates the global…

Large Language Model (LLM) based agents integrated into web browsers (often called agentic AI browsers) offer powerful automation of web tasks. However, they are vulnerable to indirect prompt injection attacks, where malicious instructions…

The growing popularity of Machine Learning (ML) has led to its deployment in various sensitive domains, which has resulted in significant research focused on ML security and privacy. However, in some applications, such as Augmented/Virtual…

Multi-robot swarms play an essential role in complex missions including battlefield reconnaissance, agricultural pest monitoring, as well as disaster search and rescue. Unfortunately, given the complexity of swarm algorithms, logical…

Private data, being larger and quality-higher than public data, can greatly improve large language models (LLM). However, due to privacy concerns, this data is often dispersed in multiple silos, making its secure utilization for LLM…

Large language models (LLMs) have demonstrated impressive capabilities across diverse tasks, but they remain susceptible to hallucinations--generating content that appears plausible but contains factual inaccuracies. We present Finch-Zk, a…

In the modern era where software plays a pivotal role, software security and vulnerability analysis are essential for secure software development. Fuzzing test, as an efficient and traditional software testing method, has been widely…

With the wide use of Deep Learning (DL) systems, academy and industry begin to pay attention to their quality. Testing is one of the major methods of quality assurance. However, existing testing techniques focus on the quality of DL models…

Fuzzing has become a popular technique for automatically detecting vulnerabilities and bugs by generating unexpected inputs. In recent years, the fuzzing process has been integrated into continuous integration workflows (i.e., continuous…

Guard models are widely used to detect harmful content in user prompts and LLM responses. However, state-of-the-art guard models rely solely on terminal-layer representations and overlook the rich safety-relevant features distributed across…

Microarchitectural attacks represent a challenging and persistent threat to modern processors, exploiting inherent design vulnerabilities in processors to leak sensitive information or compromise systems. Of particular concern is the…

The Resource Public Key Infrastructure (RPKI) has become essential to secure inter-domain routing. Despite its critical role, RPKI software remains largely untested beyond shallow parsing. Existing fuzzers, like AFL++ or libFuzzer, do not…

With the widespread application of Large Language Models across various domains, their security issues have increasingly garnered significant attention from both academic and industrial communities. This study conducts sampling and…

Large Language Models (LLMs) are increasingly served on shared accelerators where an adversary with read access to device memory can observe KV caches and hidden states, threatening prompt privacy for open-source models. Cryptographic…

WebAssembly binaries are often compiled from memory-unsafe languages, such as C and C++. Because of WebAssembly's linear memory and missing protection features, e.g., stack canaries, source-level memory vulnerabilities are exploitable in…

Recent large language models (LLMs) are rapidly extending their context windows, yet inference throughput lags due to increasing GPU memory and bandwidth demands. This is because the key-value (KV) cache, an intermediate structure storing…