Related papers: ChainCaps: Composition-Safe Tool-Using Agents via …
Production agent frameworks (OpenAI Function Calling, Anthropic Tool Use, MCP) transmit tool schemas as JSON, a format designed for machine parsing, not for interpretation by language models. For small models (4B-14B), this protocol…
As large language models (LLMs) become high-privilege agents in risk-sensitive settings, they introduce systemic threats beyond hallucination, where minor compliance errors can cause critical data leaks. However, existing benchmarks focus…
Industrial asset operations workflows are latency-sensitive because a single user query may require coordination over sensor data, work orders, failure modes, forecasting tools, and domain-specific agents. We evaluate this problem on…
Recent Pwn2Own competitions have demonstrated the continued effectiveness of control hijacking attacks despite deployed countermeasures including stack canaries and ASLR. A powerful defense called Control flow Integrity (CFI) offers a…
General-purpose agents perform tasks in unfamiliar environments without domain-specific manual customization. Yet no study has systematically measured how agent architecture shapes performance across heterogeneous protocols and diverse…
While pre-trained multimodal representations (e.g., CLIP) have shown impressive capabilities, they exhibit significant compositional vulnerabilities leading to counterintuitive judgments. We introduce Multimodal Adversarial Compositionality…
Unsafe memory accesses in programs written using popular programming languages like C/C++ have been among the leading causes for software vulnerability. Prior memory safety checkers such as SoftBound enforce memory spatial safety by…
Concurrency testing is essential to improve the reliability and security of multi-threaded programs. Dynamic analysis tools, such as TSan, depend on high-quality test drivers that reach critical shared-memory interactions at runtime.…
Coding agents powered by large language models are becoming central modules of modern IDEs, helping users perform complex tasks by invoking tools. While powerful, tool invocation opens a substantial attack surface. Prior work has…
Agentic RAG systems that equip language models with dozens to hundreds of tool definitions face a critical resource conflict: tool schemas consume the same context window needed for retrieval-augmented generation. We present the first…
Real-world tool-using agents operate over long-horizon workflows with recurring structure and diverse demands, where effective behavior requires not only invoking atomic tools but also abstracting, and reusing higher-level tool…
As frontier AI models are deployed in high-stakes decision pipelines, their ability to maintain metacognitive stability (knowing what they do not know, detecting errors, seeking clarification) under adversarial pressure is a critical safety…
We propose a Capabilities-based approach for building long-lived, complex systems that have lengthy development cycles. User needs and technology evolve during these extended development periods, and thereby, inhibit a fixed…
Tool learning empowers large language models (LLMs) as agents to use external tools and extend their utility. Existing methods employ one single LLM-based agent to iteratively select and execute tools, thereafter incorporating execution…
LLM agents increasingly act by writing code, yet a split persists between the runtime that drives the agent and the code the model writes. The runtime owns the loop, context, and control flow, and the model has little say over any of them.…
Information flow analysis is a powerful technique for reasoning about the sensitive information exposed by a program during its execution. While past work has proposed information theoretic metrics (e.g., Shannon entropy, min-entropy,…
Traditional threat modeling occurs during design, but cloud deployments introduce unanticipated threats, especially multi-stage attacks chaining vulnerabilities across trust boundaries. Existing security tools analyze components in…
Integrated Circuits (ICs) are the target of diverse attacks during their lifetime. Fabrication-time attacks, such as the insertion of Hardware Trojans, can give an adversary access to privileged data and/or the means to corrupt the IC's…
The standard operational semantics of the sequential composition operator gives rise to unbounded branching and forgetfulness when transparent process expressions are put in sequence. Due to transparency, the correspondence between…
While decoupled control schemes for legged mobile manipulators have shown robustness, learning holistic whole-body control policies for tracking global end-effector poses remains fragile against Out-of-Distribution (OOD) inputs induced by…