cs.AINov 19, 2026

Decoupled Knowledge Retrieval and Compact Memory Enable Reliable Lifelong Model Editing

Authors: Yue Che, Jingze Wang, Chen Guoyi, Yujie Wu, Xuming Ran, Mingkun Xu

Organizations: The Hong Kong Polytechnic University · Guangdong Institute of Intelligence Science and Technology · OpenMIA Research · National University of Singapore

Abstract

Lifelong model editing requires continuously incorporating new facts into deployed large language models (LLMs) without expensive retraining or compromising existing capabilities. Memory-based editors offer an appealing solution by keeping the base parameters untouched, yet their practical use is hindered by linearly growing storage requirements and deteriorating retrieval performance as edits accumulate. To tackle these problems, we introduce BiSpace, a lifelong model editing framework that decouples memory retrieval from knowledge injection across two distinct representation spaces. BiSpace routes queries and identifies relevant edits in an external semantic space, while injecting the corresponding knowledge into the internal representation space of the LLM through sparse and storage-efficient updates. Extensive experiments show that BiSpace shrinks each entry's storage to only about 10.2% of the original and shows strong resilience to long edit sequences of up to 16K edits. On a more challenging benchmark, BiSpace further improves Overall Performance by more than 23% over existing methods. These results advance memory-based editing toward reliable long-term model maintenance at scale.

Figures & tables

Appendix figures & tables8 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Towards Scalable Lifelong Knowledge Editing with Selective Knowledge Suppression

    Apr 21, 2026Dahyun Jung, Jaewook Lee, Heuiseok LimKnowledge EditingContinual Learning for LLMs

  2. LOKI: Memory-Free Null-Space Constrained Lifelong Knowledge Editing

    Jun 18, 2026Masih Eskandar, Miquel Sirera Perelló, Stratis Ioannidis +1Knowledge EditingContinual Learning for LLMs

  3. Generalizable Lifelong Model Editing via Preference Optimization

    Sep 29, 2026Dahyun Jung, Suhyune Son, Heuiseok LimContinual Learning for LLMsContinual Learning