Dial-insight: Fine-tuning Large Language Models with High-Quality Domain-Specific Data Preventing Capability Collapse
This work addresses the challenge of maintaining generalization in LLMs when fine-tuning for specialized domains like real estate, though it appears incremental as it builds on existing fine-tuning methods with a focus on data quality.
The paper tackles the problem of fine-tuning large language models for domain-specific applications without degrading their generalization capabilities, by proposing a two-stage prompt construction method and a quality assessment framework, and demonstrates that using high-quality real estate interaction data improves domain-specific proficiency without compromising generalization.
The efficacy of large language models (LLMs) is heavily dependent on the quality of the underlying data, particularly within specialized domains. A common challenge when fine-tuning LLMs for domain-specific applications is the potential degradation of the model's generalization capabilities. To address these issues, we propose a two-stage approach for the construction of production prompts designed to yield high-quality data. This method involves the generation of a diverse array of prompts that encompass a broad spectrum of tasks and exhibit a rich variety of expressions. Furthermore, we introduce a cost-effective, multi-dimensional quality assessment framework to ensure the integrity of the generated labeling data. Utilizing a dataset comprised of service provider and customer interactions from the real estate sector, we demonstrate a positive correlation between data quality and model performance. Notably, our findings indicate that the domain-specific proficiency of general LLMs can be enhanced through fine-tuning with data produced via our proposed method, without compromising their overall generalization abilities, even when exclusively domain-specific data is employed for fine-tuning.