6 papers
Reversing the Lens: Using Explainable AI to Understand Human Expertise
Roussel Rahman, Aashwin Ananda Mishra, Wan-Lin Hu
Both humans and machine learning models learn from experience, particularly in safety- and reliability-critical domains. While psychology seeks to understand human cognition, the f…
A Small Math Model: Recasting Strategy Choice Theory in an LLM-Inspired Architecture
Roussel Rahman, Jeff Shrager
Strategy Choice Theory (SCT; Siegler and Shrager, 1984; Siegler, 2000) explains important aspects of children's arithmetic learning based upon principles including learning from de…
A Fragile Number Sense: Probing the Elemental Limits of Numerical Reasoning in LLMs
Roussel Rahman, Aashwin Ananda Mishra
Large Language Models (LLMs) have demonstrated remarkable emergent capabilities, yet the robustness of their numerical reasoning remains an open question. While standard benchmarks…
Large Language Models in Numberland: A Quick Test of Their Numerical Reasoning Abilities
Roussel Rahman
An essential element of human mathematical reasoning is our number sense -- an abstract understanding of numbers and their relationships -- which allows us to solve problems involv…
Beyond Group Means and Into the World of Individuals: A Distributional Spotlight for Experimental Effects on Individuals
Roussel Rahman
Traditionally, experimental effects on humans are investigated at the group level. In this work, we present a distributional ``spotlight'' to investigate experimental effects at th…
Network Models of Expertise in the Complex Task of Operating Particle Accelerators
Roussel Rahman, Jane Shtalenkova, Aashwin Ananda Mishra +1
We implement a network-based approach to study expertise in a complex real-world task: operating particle accelerators. Most real-world tasks we learn and perform (e.g., driving ca…