StruEdit: Structured Outputs Enable the Fast and Accurate Knowledge Editing for Large Language Models
This addresses the problem of maintaining up-to-date knowledge in LLMs for question-answering systems, representing an incremental improvement over existing knowledge editing techniques.
The paper tackles the challenge of updating outdated knowledge in large language models by proposing StruEdit, which prompts models to produce structured reasoning triplets for efficient editing, achieving the highest accuracy and lowest latency compared to other methods.
As the modern tool of choice for question answering, large language models (LLMs) are expected to deliver answers with up-to-date knowledge. To achieve such ideal question-answering systems, locating and then editing outdated knowledge in the natural language outputs is a general target of popular knowledge editing methods. However, this target is challenging, as both identifying which tokens to edit in the reasoning steps and ensuring the coherence of the revised reasoning chain are difficult tasks. We argue that these challenges stem from the unstructured nature of natural language outputs. To address the above challenges, we propose $\textbf{Stru}$ctural $\textbf{Edit}$ing ($\textbf{StruEdit}$), an improved baseline for knowledge editing. We first prompt LLMs to produce structured outputs consisting of reasoning triplets. Then, StruEdit removes any potentially outdated knowledge and efficiently refills the structured outputs with up-to-date information in a single step. Experimental results show that StruEdit consistently delivers the highest accuracy with lowest latency compared with other knowledge editing methods.