Beyond Self-Consistency: Ensemble Reasoning Boosts Consistency and Accuracy of LLMs in Cancer Staging
This work addresses the need for reliable and trustworthy LLM applications in healthcare, specifically for extracting cancer staging from pathology reports, though it is incremental as it builds on existing prompting techniques like self-consistency.
The study tackled the problem of inconsistent and inaccurate cancer staging extraction from unstructured clinical reports using LLMs, and found that an ensemble reasoning approach improved both consistency and performance, achieving a 15% increase in accuracy and a 20% reduction in inconsistency compared to self-consistency methods.
Advances in large language models (LLMs) have encouraged their adoption in the healthcare domain where vital clinical information is often contained in unstructured notes. Cancer staging status is available in clinical reports, but it requires natural language processing to extract the status from the unstructured text. With the advance in clinical-oriented LLMs, it is promising to extract such status without extensive efforts in training the algorithms. Prompting approaches of the pre-trained LLMs that elicit a model's reasoning process, such as chain-of-thought, may help to improve the trustworthiness of the generated responses. Using self-consistency further improves model performance, but often results in inconsistent generations across the multiple reasoning paths. In this study, we propose an ensemble reasoning approach with the aim of improving the consistency of the model generations. Using an open access clinical large language model to determine the pathologic cancer stage from real-world pathology reports, we show that the ensemble reasoning approach is able to improve both the consistency and performance of the LLM in determining cancer stage, thereby demonstrating the potential to use these models in clinical or other domains where reliability and trustworthiness are critical.