Leveraging LLMs for Title and Abstract Screening for Systematic Review: A Cost-Effective Dynamic Few-Shot Learning Approach

TL;DR

This paper introduces a two-stage dynamic few-shot learning method utilizing large language models to improve the efficiency and accuracy of title and abstract screening in systematic reviews, significantly reducing manual effort and costs.

Contribution

The paper presents a novel two-stage approach that combines low-cost and high-performance LLMs for cost-effective, accurate screening in systematic reviews.

Findings

Reduces manual screening effort by up to 50%

Demonstrates strong generalizability across 10 reviews

Achieves cost savings while maintaining high accuracy

Abstract

Systematic reviews are a key component of evidence-based medicine, playing a critical role in synthesizing existing research evidence and guiding clinical decisions. However, with the rapid growth of research publications, conducting systematic reviews has become increasingly burdensome, with title and abstract screening being one of the most time-consuming and resource-intensive steps. To mitigate this issue, we designed a two-stage dynamic few-shot learning (DFSL) approach aimed at improving the efficiency and performance of large language models (LLMs) in the title and abstract screening task. Specifically, this approach first uses a low-cost LLM for initial screening, then re-evaluates low-confidence instances using a high-performance LLM, thereby enhancing screening performance while controlling computational costs. We evaluated this approach across 10 systematic reviews, and the results demonstrate its strong generalizability and cost-effectiveness, with potential to reduce manual screening burden and accelerate the systematic review process in practical applications.