Editing conceptual knowledge for large language models

X Wang, S Mao, N Zhang, S Deng, Y Yao… - arXiv preprint arXiv …, 2024 - arxiv.org
Recently, there has been a growing interest in knowledge editing for Large Language
Models (LLMs). Current approaches and evaluations merely explore the instance-level …

Detoxifying Large Language Models via Knowledge Editing

M Wang, N Zhang, Z Xu, Z Xi, S Deng, Y Yao… - arXiv preprint arXiv …, 2024 - arxiv.org
This paper investigates using knowledge editing techniques to detoxify Large Language
Models (LLMs). We construct a benchmark, SafeEdit, which covers nine unsafe categories …

How Well Can Knowledge Edit Methods Edit Perplexing Knowledge?

H Ge, F Rudzicz, Z Zhu - arXiv preprint arXiv:2406.17253, 2024 - arxiv.org
As large language models (LLMs) are widely deployed, targeted editing of their knowledge
has become a critical challenge. Recently, advancements in model editing techniques, such …

TAXI: Evaluating Categorical Knowledge Editing for Language Models

D Powell, W Gerych, T Hartvigsen - arXiv preprint arXiv:2404.15004, 2024 - arxiv.org
Humans rarely learn one fact in isolation. Instead, learning a new fact induces knowledge of
other facts about the world. For example, in learning a korat is a type of cat, you also infer it …

Keys to Robust Edits: from Theoretical Insights to Practical Advances

J Yan, F Wang, Y Luo, Y Li, Y Zhang - arXiv preprint arXiv:2410.09338, 2024 - arxiv.org
Large language models (LLMs) have revolutionized knowledge storage and retrieval, but
face challenges with conflicting and outdated information. Knowledge editing techniques …