English

Synthetic Knowledge Ingestion: Towards Knowledge Refinement and Injection for Enhancing Large Language Models

Computation and Language 2024-10-15 v1 Artificial Intelligence Information Retrieval Machine Learning

Abstract

Large language models (LLMs) are proficient in capturing factual knowledge across various domains. However, refining their capabilities on previously seen knowledge or integrating new knowledge from external sources remains a significant challenge. In this work, we propose a novel synthetic knowledge ingestion method called Ski, which leverages fine-grained synthesis, interleaved generation, and assemble augmentation strategies to construct high-quality data representations from raw knowledge sources. We then integrate Ski and its variations with three knowledge injection techniques: Retrieval Augmented Generation (RAG), Supervised Fine-tuning (SFT), and Continual Pre-training (CPT) to inject and refine knowledge in language models. Extensive empirical experiments are conducted on various question-answering tasks spanning finance, biomedicine, and open-generation domains to demonstrate that Ski significantly outperforms baseline methods by facilitating effective knowledge injection. We believe that our work is an important step towards enhancing the factual accuracy of LLM outputs by refining knowledge representation and injection capabilities.

Keywords

Cite

@article{arxiv.2410.09629,
  title  = {Synthetic Knowledge Ingestion: Towards Knowledge Refinement and Injection for Enhancing Large Language Models},
  author = {Jiaxin Zhang and Wendi Cui and Yiran Huang and Kamalika Das and Sricharan Kumar},
  journal= {arXiv preprint arXiv:2410.09629},
  year   = {2024}
}

Comments

EMNLP 2024 main conference long paper

R2 v1 2026-06-28T19:19:10.490Z