Work

Version 2.0 - EmbedSLR: An open-source python framework for efficient embedding-based screening and bibliometric validation in systematic literature review

Year: 2026

Type: software-paper

Abstract: EmbedSLR 2.0 features a multi-embedding agent system that simulates teamwork - a vital aspect of systematic literature reviews (SLR) and evaluates it using bibliometric metrics. Experiments demonstrate that the models' effectiveness varies significantly and that publications selected by consensus outperform those chosen by individual models. EmbedSLR 2.0 allows users to choose the best model for a specific research topic or further enhance the matching of the selected publication to the research subject through model consensus. Building on v1.0, this version further improves effectiveness, introduces results visualization, supports both local and Colab environments, ensures reproducibility, and remains easy to use.

Source: SoftwareX

DOI: 10.1016/j.softx.2026.102563

Authors: Sebastian Matysik, Joanna Wiśniewska, Paweł Karol Frankowski

Institutions: University of Szczecin, West Pomeranian University of Technology in Szczecin, Maritime University of Szczecin

Language: English

Topic: Meta-analysis and systematic reviews

Subfield: Statistics, Probability and Uncertainty

Field: Decision Sciences

Domain: Social Sciences

Open Access status: gold

Estimated APC paid: $1,200

Funder: Narodowe Centrum Nauki