Instruction-tuned Language Models are Better Knowledge Learners

arXiv:2402.12847v1 Announce Type: cross
Abstract: In order for large language model (LLM)-based assistants to effectively adapt to evolving information needs, it must be possible to update their factual knowledge through continued training on new data. The standard recipe for doing so involves continued pre-training on new documents followed by instruction-tuning on question-answer (QA) pairs. However, we find that LLMs trained with this recipe struggle to answer questions, even though the perplexity of documents is minimized. We found that QA pairs …

abstract adapt arxiv assistants cs.ai cs.cl cs.lg data documents information instruction-tuned knowledge language language model language models large language large language model llm pre-training question recipe standard through training type update