Uncertainty Under the Curve: A Sequence-Level Entropy Area Metric for Reasoning LLM
This provides a practical tool for uncertainty modeling and data quality assessment in LLM training, though it is incremental as it builds on existing entropy-based methods.
The paper tackles the problem of quantifying uncertainty in reasoning large language models by introducing the Entropy Area Score (EAS), a metric that integrates token-level predictive entropy without external models or repeated sampling. Results show EAS is strongly correlated with answer entropy and outperforms Pass Rate filtering in training data selection, improving student model accuracy on math benchmarks.
In this work, we introduce Entropy Area Score (EAS), a simple yet effective metric to quantify uncertainty in the answer generation process of reasoning large language models (LLMs). EAS requires neither external models nor repeated sampling, it integrates token-level predictive entropy from the model itself to capture the evolution of uncertainty during generation. Empirical results show that EAS is strongly correlated with answer entropy across models and datasets. In training data selection, EAS identifies high-potential samples and consistently outperforms Pass Rate filtering under equal sample budgets, improving student model accuracy on math benchmarks. EAS is both efficient and interpretable, offering a practical tool for uncertainty modeling and data quality assessment in LLM training.