Job Skill Extraction via LLM-Centric Multi-Module Framework

TL;DR

This paper introduces SRICL, a comprehensive framework that enhances large language model-based skill extraction from job ads by combining semantic retrieval, in-context learning, supervised fine-tuning, and verification, improving accuracy and reliability.

Contribution

The paper presents SRICL, a novel multi-module framework that significantly improves span-level skill extraction accuracy and robustness across diverse domains and languages.

Findings

SRICL outperforms GPT-3.5 prompting baselines in STRICT-F1 scores.

It reduces invalid tags and hallucinations in extracted spans.

Enables dependable deployment in low-resource, multi-domain settings.

Abstract

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.