One for All: Update Parameterized Knowledge Across Multiple Models

TL;DR

OnceEdit is a novel ensemble-based method that enables efficient and stable knowledge updates across multiple large language models by using a plug-in editing module and dynamic weighting mechanisms.

Contribution

The paper introduces OnceEdit, a new ensemble-based approach for multi-model knowledge editing, addressing limitations of existing methods focused on single models.

Findings

Outperforms existing knowledge editing methods in efficiency and effectiveness.

Demonstrates stable and adaptable knowledge updates across diverse large language models.

Effective in multi-model editing scenarios with improved stability.

Abstract

Large language models (LLMs) encode vast world knowledge but struggle to stay up-to-date, often leading to errors and hallucinations. Knowledge editing offers an efficient alternative to retraining, enabling targeted modifications by updating specific model parameters. However, existing methods primarily focus on individual models, posing challenges in efficiently updating multiple models and adapting to new models. To address this, we propose OnceEdit, a novel ensemble-based approach that employs a plug-in model as the editing module, enabling stable knowledge updates across multiple models. Building on the model ensemble, OnceEdit introduces two key mechanisms to enhance its effectiveness. First, we introduce a dynamic weight mechanism through a \weight token for distinguishing between edit-related and non-edit-related instances, ensuring the appropriate utilization of knowledge from integrated models. Second, we incorporate an ensemble enhancement mechanism to mitigate the excessive reliance on the central model inherent in the model ensemble technique, making it more suitable for knowledge editing. Extensive experiments on diverse LLMs demonstrate that OnceEdit consistently outperforms existing methods while achieving superior editing efficiency. Further analysis confirms its adaptability and stability in multi-model editing scenarios. Our code will be available.