Related papers: When Should Users Check? Modeling Confirmation Fre…
To enable human oversight, agentic AI systems often provide a trace of reasoning and action steps. Designing traces to have an informative, but not overwhelming, level of detail remains a critical challenge. In three user studies on a…
Background: Large-scale biological jobs on high-performance computing systems require manual intervention if one or more computing cores on which they execute fail. This places not only a cost on the maintenance of the job, but also a cost…
Scalable oversight protocols aim to empower evaluators to accurately verify AI models more capable than themselves. However, human evaluators are subject to biases that can lead to systematic errors. We conduct two studies examining the…
Artificial intelligence (AI)-based decision support systems hold promise for enhancing diagnostic accuracy and efficiency in computational pathology. However, human-AI collaboration can introduce and amplify cognitive biases, such as…
As autonomous agents become increasingly sophisticated, validating their sequential behavior presents a significant challenge. Traditional testing approaches require manual specification, exact sequence matching, or thousands of training…
Large language model (LLM)-based agents are increasingly used to perform complex, multi-step workflows in regulated settings such as compliance and due diligence. However, many agentic architectures rely primarily on prompt engineering of a…
AI agents are increasingly used to solve complex, multi-step tasks, but existing multi-agent frameworks remain brittle as workflows grow in scale and depth. Small errors at intermediate stages can propagate through agent interactions, while…
Recent search agents leverage multi-turn reasoning and search tools to achieve strong performance on multi-hop and long-horizon benchmarks. Yet it remains unclear whether they reliably reason across all requirements by tracking, verifying,…
Recent work has shown that, in classification tasks, it is possible to design decision support systems that do not require human experts to understand when to cede agency to a classifier or when to exercise their own agency to achieve…
Saving, or checkpointing, intermediate results during interactive data exploration can potentially boost user productivity. However, existing studies on this topic are limited, as they primarily rely on small-scale experiments with human…
People are increasingly turning to AI assistance for simple tasks, e.g., arithmetic, spell-check, and answering simple questions. But does AI assistance actually save users time and effort? We investigate people's propensity to use AI for…
Reducing the number of failures in a production system is one of the most challenging problems in technology driven industries, such as, the online retail industry. To address this challenge, change management has emerged as a promising…
Quantifying the workplace productivity effects of Generative Artificial Intelligence is now central to economics, management, and public policy. The deployment of AI tools in customer service, writing, software development, and consulting…
In many real-world continuous action domains, human agents must decide which actions to attempt and then execute those actions to the best of their ability. However, humans cannot execute actions without error. Human performance in these…
Large Reasoning Models (LRMs) achieve strong performance by generating long reasoning traces with reflection. Through a large-scale empirical analysis, we find that a substantial fraction of reflective steps consist of self-verification…
Runtime verification consists in observing and collecting the execution traces of a system and checking them against a specification, with the objective of raising an error when a trace does not satisfy the specification. We consider…
Tool-calling agents are evaluated on tool selection, parameter accuracy, and scope recognition, yet LLM trajectory assessments remain inherently post-hoc. Disconnected from the active execution loop, such assessments identify errors that…
Corrections offer a natural modality for people to provide feedback to a robot, by (i) intervening in the robot's behavior when they believe the robot is failing (or will fail) the task objectives and (ii) modifying the robot's behavior to…
Multi-Agent Systems (MAS) are notoriously complex and hard to verify. In fact, it is not trivial to model a MAS, and even when a model is built, it is not always possible to verify, in a formal way, that it is actually behaving as we…
Evaluating human-AI decision-making systems is an emerging challenge as new ways of combining multiple AI models towards a specific goal are proposed every day. As humans interact with AI in decision-making systems, multiple factors may be…