Self-Guided Framework for Improving Arithmetic Reasoning in Large Language Models with Reinforcement Learning
Status: CompletedAuthors: Jiwoo Hong, Huzama Ahmad, Minsu Kim, James Thorne
Can a language model get better at math by checking its own work? This project starts by measuring how reliably models can evaluate themselves, then turns that signal into training: a reinforcement-learning loop that rewards the model's greedy answer when it holds up against its own diverse samples, so the reasoning sharpens without a human grader in the loop. Across models from one to twenty billion parameters, it lifts accuracy by up to 5% on four math-reasoning benchmarks while also improving commonsense and holding onto general language ability, and both human and automated review found the explanations grew more detailed and logical.