Span-level skill extraction from job advertisements underpins candidate-job matching and labor-market analytics, yet generative large language models (LLMs) often yield malformed spans, boundary drift, and hallucinations, especially with long-tail terms and cross-domain shift. We present SRICL, an LLM-centric framework that combines semantic retrieval (SR), in-context learning (ICL), and supervised fine-tuning (SFT) with a deterministic verifier. SR pulls in-domain annotated sentences and definitions from ESCO to form format-constrained prompts that stabilize boundaries and handle coordination. SFT aligns output behavior, while the verifier enforces pairing, non-overlap, and BIO legality with minimal retries. On six public span-labeled corpora of job-ad sentences across sectors and languages, SRICL achieves substantial STRICT-F1 improvements over GPT-3.5 prompting baselines and sharply reduces invalid tags and hallucinated spans, enabling dependable sentence-level deployment in low-resource, multi-domain settings.
@article{arxiv.2604.21525,
title = {Job Skill Extraction via LLM-Centric Multi-Module Framework},
author = {Guojing Li and Zichuan Fu and Junyi Li and Faxue Liu and Wenxia Zhou and Yejing Wang and Jingtong Gao and Maolin Wang and Rungen Liu and Wenlin Zhang and Xiangyu Zhao},
journal= {arXiv preprint arXiv:2604.21525},
year = {2026}
}