LyapLock: Bounded Knowledge Preservation in Sequential Large Language Model Editing
This addresses the issue of long-term knowledge preservation in LLMs for users needing reliable model updates, representing a novel method rather than an incremental improvement.
The paper tackles the problem of progressive performance decline in sequential large language model editing by proposing LyapLock, a framework that integrates queuing theory and Lyapunov optimization to achieve asymptotic optimal editing performance, scaling to over 10,000 edits with an 11.89% boost in average editing efficacy over SOTA baselines.
Large Language Models often contain factually incorrect or outdated knowledge, giving rise to model editing methods for precise knowledge updates. However, current mainstream locate-then-edit approaches exhibit a progressive performance decline during sequential editing, due to inadequate mechanisms for long-term knowledge preservation. To tackle this, we model the sequential editing as a constrained stochastic programming. Given the challenges posed by the cumulative preservation error constraint and the gradually revealed editing tasks, \textbf{LyapLock} is proposed. It integrates queuing theory and Lyapunov optimization to decompose the long-term constrained programming into tractable stepwise subproblems for efficient solving. This is the first model editing framework with rigorous theoretical guarantees, achieving asymptotic optimal editing performance while meeting the constraints of long-term knowledge preservation. Experimental results show that our framework scales sequential editing capacity to over 10,000 edits while stabilizing general capabilities and boosting average editing efficacy by 11.89\% over SOTA baselines. Furthermore, it can be leveraged to enhance the performance of baseline methods. Our code is released on https://github.com/caskcsg/LyapLock.