Shashank Sonkar

This study investigates how often language models reproduce exact text from training data.

MalAlgoQA dataset evaluates reasoning of Large Language Models in counterfactual scenarios.

A detailed method improves physics learning and assessment for students and educators.