Related papers: Trace-Level Analysis of Information Contamination …

In Agentic AI, Large Language Models (LLMs) are increasingly used in the orchestration layer to coordinate multiple agents and to interact with external services, retrieval components, and shared memory. In this setting, failures are not…

Agentic systems have recently emerged as a promising tool to automate literature-based ideation. However, current systems often remain black-box, with limited transparency or control for researchers. Our work introduces TrustResearcher, a…

Multi-agent systems built on large language models (LLMs) require many coordination choices that are difficult to fix a priori: which skill protocol to invoke, which agent role should perform a subtask, which model to bind to each role, how…

Conversational agents are increasingly deployed in knowledge-intensive settings, where correct behavior depends on retrieving and applying domain-specific knowledge from large, proprietary, and unstructured corpora during live interactions…

Data contamination is a known threat to the reliability of model evaluation. However, it remains underexplored in code large language models (LLMs), where contamination often goes beyond exact duplication. We present TRACER, a…

Traditional models of market efficiency assume that equity prices incorporate information based on content alone, often neglecting the structural influence of reporting timing and cadence. This study introduces the Autonomous Disclosure…

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,…

Code agents are advancing rapidly, but debugging them is becoming increasingly difficult. As frameworks orchestrate parallel tool calls and multi-stage workflows over complex tasks, making the agent's state transitions and error propagation…

Agentic systems are becoming more capable: agents define strategies, take actions, and interact with different environments. This autonomy poses serious challenges for overseeing and assessing agent behavior. Most current tools are limited,…

Retrieval-Augmented Generation (RAG) systems are vulnerable to knowledge base poisoning, yet existing attacks have been evaluated almost exclusively against vanilla retrieve-then-generate pipelines. Architectures designed to handle…

Complex networked systems can be modeled and represented as graphs, with nodes representing the agents and the links describing the dynamic coupling between them. The fundamental objective of network identification for dynamic systems is to…

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…

Evaluations of generative AI models often collapse nuanced behaviour into a single number computed for a single decoding configuration. Such point estimates obscure tail risks, demographic disparities, and the existence of multiple…

Background: Large language models are typically evaluated as models, benchmarks, or short conversational episodes. Less is known about what happens when an agent is embedded persistently in a real academic research environment with durable…

Unpredictable behavior is often taken as evidence of control, yet stochastic dispersion and structured action control need not coincide. This paper tests whether stochastic sampling can substitute for structured mechanisms that couple…

We consider the automatic verification of information flow security policies of web-based workflows, such as conference submission systems like EasyChair. Our workflow description language allows for loops, non-deterministic choice, and an…

High-precision CNC machining of free-form aerospace components requires bounded compensations informed by inspection, simulation, and process knowledge. Off-the-shelf large language model (LLM) assistants can generate text, but they do not…

When automated decision systems fail, organizations frequently discover that formally compliant governance infrastructure cannot reconstruct what happened or why. This paper synthesizes an operational governance evidence framework --…

In-context learning enables large language models to perform novel tasks through few-shot demonstrations. However, demonstrations per se can naturally contain noise and conflicting examples, making this capability vulnerable. To understand…

Safety evaluation for advanced AI systems assumes that behavior observed under evaluation predicts behavior in deployment. This assumption weakens for agents with situational awareness, which may exploit regime leakage, cues distinguishing…