Year: 2026
Type: conference-paper
Abstract: Adapting Large Language Models (LLMs) to specialized domains requires high-quality instruction tuning datasets, which are expensive to create through human annotation.Existing data synthesis methods focus on generalpurpose tasks and fail to capture domainspecific terminology and reasoning patterns.To address this, we introduce DS 2 -INSTRUCT, a zero-shot framework that generates domainspecific instruction datasets without human supervision.Our approach first generates taskinformed keywords to ensure comprehensive domain coverage.It then creates diverse instructions by pairing these keywords with different cognitive levels from Bloom's Taxonomy.Finally, it uses self-consistency validation to ensure data quality.We apply this framework to generate datasets across seven challenging domains, such as mathematics, finance, and logical reasoning.Comprehensive evaluation demonstrates that models fine-tuned on our generated data achieve substantial improvements over existing data generation methods. 1
Authors: Ruiyao Xu, Noelle I. Samia, Han Liu
Institution: Northwestern University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7140140937