---
title: "BetaEdit: Null-Space Constrained Sequential Model Editing"
authors:
  - "Bingqing Liu"
  - "Wei Liu"
  - "Yuhua Li"
doi: "10.24963/ijcai.2026/508"
url: "https://ijcai.org/proceedings/2026/508"
pdf: "https://ijcai.org/proceedings/2026/0508.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "4563-4571"
track: "Main Track"
keywords:
  - "Natural Language Processing: Language models"
  - "Natural Language Processing: Interpretability and analysis of models for NLP"
  - "Machine Learning: Incremental learning"
---

# BetaEdit: Null-Space Constrained Sequential Model Editing

Bingqing Liu, Wei Liu, Yuhua Li

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 4563-4571. DOI: [10.24963/ijcai.2026/508](https://doi.org/10.24963/ijcai.2026/508)

## Abstract

Null-space-based methods have garnered considerable attention in model editing by constraining updates to the null space of the pre-existing knowledge representation, thereby preserving the model's original behavior. However, in practice these methods rely on an approximate null space—leading to knowledge leakage—and further suffer from severe performance degradation during sequential editing. Recent work shows that history-aware editing strategies can empirically mitigate this decline, yet the underlying reason remains unclear. In this paper, we first expose the knowledge leakage inherent in existing null-space approaches and then analyze why history-aware updates effectively preserve both editing  performance and general capabilities during long-horizon editing. Building on these insights, we propose BetaEdit, a refined framework that effectively controls the  knowledge leakage and integrates history-aware updates into the null-space paradigm. Extensive experiments on three large language models across two standard benchmarks show that BetaEdit consistently outperforms prior methods in the challenging regime of massive-scale sequential editing.  Code is available at: https://github.com/lbq8942/BetaEdit.