Evaluating Mathematical Reasoning Beyond Accuracy
This addresses the need for better evaluation of reasoning in LLMs for mathematical tasks, though it is incremental as it builds on existing evaluation frameworks.
The paper tackles the problem that current evaluations of Large Language Models (LLMs) on mathematical tasks focus only on final-answer accuracy, neglecting reasoning step quality, and introduces ReasonEval, a methodology using validity and redundancy metrics with LLM-based evaluators, which outperforms baselines in meta-evaluation datasets and shows strong generalization.
The leaderboard of Large Language Models (LLMs) in mathematical tasks has been continuously updated. However, the majority of evaluations focus solely on the final results, neglecting the quality of the intermediate steps. This oversight can mask underlying problems, such as logical errors or unnecessary steps in the reasoning process. To measure reasoning beyond final-answer accuracy, we introduce ReasonEval, a new methodology for evaluating the quality of reasoning steps. ReasonEval employs validity and redundancy to characterize the reasoning quality, as well as accompanying LLMs to assess them automatically. We explore different design options for the LLM-based evaluators and empirically demonstrate that ReasonEval, when instantiated with base models possessing strong mathematical knowledge and trained with high-quality labeled data, consistently outperforms baseline methods in the meta-evaluation datasets. We also highlight the strong generalization capabilities of ReasonEval. By utilizing ReasonEval to evaluate LLMs specialized in math, we find that an increase in final-answer accuracy does not necessarily guarantee an improvement in the overall quality of the reasoning steps for challenging mathematical problems. Additionally, we observe that ReasonEval can play a significant role in data selection. We open-source the best-performing model, meta-evaluation script, and all evaluation results to facilitate future research.