Efficient Text Encoders for Labor Market Analysis
Open Access
- 14 July 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 133596-133608
Abstract
Labor market analysis relies on extracting insights from job advertisements, which provide valuable yet unstructured information on job titles and corresponding skill requirements. While state-of-the-art methods for skill extraction achieve strong performance, they depend on large language models (LLMs), which are computationally expensive and slow. In this paper, we propose ConTeXT-match, a novel contrastive learning approach with token-level attention that is well-suited for the extreme multi-label classification task of skill classification. ConTeXT-match significantly improves skill extraction efficiency and performance, achieving state-of-the-art results with a lightweight bi-encoder model. To support robust evaluation, we introduce Skill-XL a new benchmark with exhaustive, sentence-level skill annotations that explicitly address the redundancy in the large label space. Finally, we present JobBERT V2, an improved job title normalization model that leverages extracted skills to produce high-quality job title representations. Experiments demonstrate that our models are efficient, accurate, and scalable, making them ideal for large-scale, real-time labor market analysis.Keywords
All Related Versions
Funding Information
- Flemish Government, through Flanders Innovation and Entrepreneurship (HBC.2020.2893)
- “Onderzoeksprogramma Artificiële Intelligentie (AI) Vlaanderen” Program;
This publication has 22 references indexed in Scilit:
- Data science for job market analysis: A survey on applications and techniquesExpert Systems with Applications, 2024
- Automated occupation coding with hierarchical features: a data-centric approach to classification with pre-trained language modelsDiscover Artificial Intelligence, 2023
- A Survey on Skill Identification From Online Job AdsIEEE Access, 2021
- DeepCarotene -Job Title Classification with Multi-stream Convolutional Neural Network2019
- Learning Representations for Soft Skill Matching2018
- Representation of Job-Skill in Artificial Intelligence with Knowledge Graph Analysis2018
- Employer Voices, Employer Demands, and Implications for Public Skills Development Policy Connecting the Labor and Education SectorsThe World Bank Research Observer, 2016
- Carotene: A Job Title Classification System for the Online Recruitment Domain2015
- High-precision phrase-based document classification on a modern scale2011
- Dimensionality Reduction by Learning an Invariant Mapping2006