9 papers
When Should Users Check? Modeling Confirmation Frequency inMulti-Step Agentic AI Tasks
Jieyu Zhou, Aryan Roy, Sneh Gupta +2
Existing AI agents typically execute multi-step tasks autonomously and only allow user confirmation at the end. During execution, users have little control, making the confirm-at-e…
Guidelines for Designing AI Technologies to Support Adult Learning
Jennifer M. Reddig, Glen R. Smith, Sanaz Ahmadzadeh Siyahrood +16
AI-powered educational technologies have demonstrated measurable benefits for learners, but their design and evaluation have largely centered on K-12 contexts. As a result, many AI…
STAND: Self-Aware Precondition Induction for Interactive Task Learning
Daniel Weitekamp, Glen Smith, Kenneth Koedinger +1
In interactive task learning (ITL), AI agents learn new capabilities from limited human instruction provided during task execution. STAND is a new method of data-efficient rule pre…
CORGI: Efficient Pattern Matching With Quadratic Guarantees
Daniel Weitekamp
Rule-based systems must solve complex matching problems within tight time constraints to be effective in real-time applications, such as planning and reactive control for AI agents…
Decomposed Inductive Procedure Learning: Learning Academic Tasks with Human-Like Data Efficiency
Daniel Weitekamp, Christopher MacLellan, Erik Harpstead +1
Human learning relies on specialization -- distinct cognitive mechanisms working together to enable rapid learning. In contrast, most modern neural networks rely on a single mechan…
TutorGym: A Testbed for Evaluating AI Agents as Tutors and Students
Daniel Weitekamp, Momin N. Siddiqui, Christopher J. MacLellan
Recent improvements in large language model (LLM) performance on academic benchmarks, such as MATH and GSM8K, have emboldened their use as standalone tutors and as simulations of h…