2 papers
cs.CL2026
Style-Debiased DPO: Updating LLM Knowledge with Factuality-Aware Synthetic Preference Data
Takayuki Yamamoto, Daisuke Kawahara
Continued pretraining (CPT) with data augmentation such as paraphrasing can store inside a large language model (LLM) the knowledge of a small source corpus. The stored knowledge,…
cs.CL2026
PASTA: A Paraphrasing And Self-Training Approach for Knowledge Updating in LLMs
Takayuki Yamamoto, Daisuke Kawahara
Knowledge updating in pre-trained Large Language Models (LLMs) remains an important challenge. While continual training provides a potential avenue for knowledge updating, it conti…