Modular Refinement of Small Language Models for Physics Reasoning via Localized Error Feedback
Lists
Only you see this
Abstract
Large Language Models (LLMs) excel at many reasoning tasks but struggle with scientific domains like physics, which demand precise mathematical calculations alongside deep conceptual and factual understanding. In complex physics problem solving, LLMs commonly falter due to three core issues: misunderstanding the problem, incorrect application of concepts, and calculation mistakes. These challenges are more pronounced in small LLMs due to their limited capacity, making them more prone to failures. To address these limitations, we propose a modular reinforcement learning refinement framework tailored for small LLMs, integrating first step error localization, and correction through a Reinforcement Learning guided feedback mechanism. We also introduce PhysicsQA, a diverse benchmark of 370 physics problems designed to evaluate LLM reasoning across the aforementioned dimensions. Experimental results demonstrate improvements upto 10% in final answer accuracy reasoning using Small language models over existing approaches
BibTeX
@misc{jaiswal2026modular,
author = {Raj Jaiswal and Dhruv Jain and Rishabh Dhawan and Dhruvkumar Patel and Avinash Anand and Shin'ichi Satoh and Tanuja Ganu and Rajiv Ratn Shah and Erik Cambria and Zhengkui Wang},
title = {{Modular Refinement of Small Language Models for Physics Reasoning via Localized Error Feedback}},
year = {2026},
url = {https://openreview.net/forum?id=DiQ8d2kOxG},
howpublished = {OpenReview},
note = {Submitted to ICLR 2026},
}Loading related content...