A Unified Framework for Model Editing

arXiv:2403.14236v1 Announce Type: new
Abstract: Model editing is a growing area focused on updating the knowledge embedded within models. Among the various methodologies, ROME and MEMIT stand out as leading "locate-and-edit" model editing techniques. While MEMIT enables batched editing of memories, ROME is limited to changing one fact at a time. This paper introduces a unifying framework that brings ROME and MEMIT under a single conceptual umbrella, optimizing for the same goal, which we call the "preservation-memorization" objective. This objective …

abstract arxiv cs.ai cs.cl cs.lg edit editing embedded framework knowledge memories paper type