Enhancing Job Recommendation through LLM-Based Generative Adversarial Networks

Abstract

Recommending suitable jobs to users is a critical task in online recruitment platforms. Existing job recommendation methods often encounter challenges such as the low quality of users' resumes, which hampers their accuracy and practical effectiveness. With the rapid development of large language models (LLMs), utilizing the rich knowledge encapsulated within them, as well as their powerful reasoning capabilities, offers a promising avenue for enhancing resume completeness to achieve more accurate recommendations. However, directly leveraging LLMs is not a one-size-fits-all solution, as it may suffer from issues like fabricated generation and few-shot problem, both of which can degrade the quality of resume completion. In this paper, we propose a novel LLM-based GANs Interactive Recommendation (LGIR) approach for job recommendation. To alleviate the limitation of fabricated generation, we not only extract users' explicit properties (e.g., skills, interests) from their self-description but also infer users' implicit characteristics from their behaviors for more accurate and meaningful resume completion. Nevertheless, some users still suffer from the few-shot problem, which arises due to scarce interaction records, leading to limited guidance for high-quality resume generation. To address this issue, we propose aligning unpaired low-quality resumes with high-quality generated counterparts using Generative Adversarial Networks (GANs), which can refine resume representations for better recommendation results. Extensive experiments on three large real-world recruitment datasets demonstrate the effectiveness of our proposed method.