LLM4SCREENLIT: Recommendations on Assessing the Performance of Large Language Models for Screening Literature in Systematic Reviews

TL;DR

This paper introduces WMCC, a new metric for evaluating large language models in literature screening, emphasizing cost-sensitive assessment and comprehensive reporting practices.

Contribution

It proposes WMCC, a cost-sensitive evaluation metric, and provides practical guidelines for assessing LLM performance in systematic review screening.

Findings

WMCC often disagrees with accuracy and MCC in ranking LLMs.

Most studies lack full confusion matrix reporting and cost-sensitive metrics.

Using WMCC reduces false negatives in literature screening evaluations.

Abstract

Context: Large language models (LLMs) are increasingly used to screen literature for systematic reviews (SRs), but the standard confusion-matrix metrics used to evaluate them can mislead under the imbalanced, cost-asymmetric conditions of screening. Objective: We develop and justify LLM4SCREENLIT-practical recommendations for researchers conducting LLM-screening evaluations and for editors and reviewers assessing such studies-differentiated by study type (retrospective benchmarking vs deployment for a specific SR). Method: Using Delgado-Chaves et al. (2025), an 18-LLM benchmark across three biomedical SRs, as a motivating example, we reviewed 28 additional papers and extracted their reported metrics. We propose a Weighted Matthews Correlation Coefficient (WMCC) that integrates MCC's chance-correction with asymmetric misclassification costs, and validated it on three software-engineering (SE) reanalyses, the largest covering 9 LLMs x 24 SE secondary studies (34,528 articles). Results: Across the 29 papers, only 10% reported MCC, only 24% reported full confusion matrices, and none of the five papers claiming workload savings priced false-negative cost. In the largest SE reanalysis, MCC and WMCC disagree on the best LLM in 55% of evaluable studies; in the most striking 9,695-article SE study, the Accuracy-best LLM loses 63.3% of relevant evidence (Lost Evidence), the MCC-best 43.9%, but the WMCC-best only 5.8%. Sensitivity analysis (median crossover at w~=2.7, all <7) supports w=10 as a conservative default. Conclusions: SR-screening evaluations should prioritize Lost Evidence and use cost-sensitive WMCC alongside MCC for ranking. Reporting must include the full confusion matrix and treat unclassifiable outputs as positives requiring human review. Designs should be leakage-aware, with non-LLM baselines when the study aims to inform SR practice and labels are available.