More Edits, More Stable: Understanding the Lifelong Normalization in Sequential Model Editing

• URL: http://arxiv.org/abs/2605.11836v1
• Date: Tue, 12 May 2026 09:23:35 GMT
• Title: More Edits, More Stable: Understanding the Lifelong Normalization in Sequential Model Editing
• Abstract summary: Lifelong Model Editing aims to continuously update evolving facts in Large Language Models while preserving unrelated knowledge and general capabilities.<n>We find that recent editors resilient over long horizons share the same core strategy: Lifelong Normalization (LN), which normalizes value using running statistics.<n>LN causes immediate performance collapse, and we observe a counter-intuitive positive cumulative effect where early edits can promote the success of future edits.
• Score: 53.41147779846785
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Lifelong Model Editing aims to continuously update evolving facts in Large Language Models while preserving unrelated knowledge and general capabilities, yet it remains plagued by catastrophic forgetting and model collapse. Empirically, we find that recent editors resilient over long horizons share the same core strategy: Lifelong Normalization (LN), which normalizes value gradients using running statistics. Removing LN causes immediate performance collapse, and we observe a counter-intuitive positive cumulative effect where early edits can promote the success of future edits. Yet the mechanism of LN remains a "black box", leaving its precise role in lifelong stability poorly understood. In this work, we provide the first theoretical account of LN in the lifelong regime. Our analysis reveals a self-reinforcing stability loop and proves that, when combined with ridge-regularized regression, LN yields parameter updates with asymptotic orthogonality and bounded norms, directly mitigating forgetting and systemic collapse. Based on these insights, we derive StableEdit, which strengthens this stability loop via an explicit warm-up stage and full whitening, improving long-horizon stability at minimal overhead. Extensive experiments validate our theory and demonstrate competitive performance. Our code is available at https://github.com/MINE-USTC/StableEdit.

Related papers

• The Road Taken: The Role of Optimizers at the Edge of Stability [53.81268610971847]
  edge of stability refers to a phenomenon in deep learning where the Hessian eigenvalues of the loss remain stable above a threshold that the classical descent predicts to be unstable.<n>Previous works formulate the edge of stability with respect to the maximum Hessian eigenvalue and the learning rate.<n>We observe that many first-order methods, including gradient descent, significantly violate the stability bound predicted by these theories by a factor as large as $times 21.1$.
  arXiv  Detail & Related papers  (2026-08-19T01:03:34Z)
• Toward Ultra-Long-Horizon Sequential Model Editing [0.43748379918040853]
  We show that long sequences of edits trigger abrupt model collapse in large language models.<n>We propose Norm-Anchor Scaling NAS, a plug-and-play norm-constrained strategy.<n>NAS delays the collapse point of representative L&E algorithms by more than 4 times and yields a 72.2% average relative gain in editing performance.
  arXiv  Detail & Related papers  (2026-01-30T04:31:21Z)
• Beyond Hard Writes and Rigid Preservation: Soft Recursive Least-Squares for Lifelong LLM Editing [16.768713543557443]
  Model editing updates a pre-trained with new facts or rules without re-training, while preserving unrelated behavior.<n>We propose RLSEdit, a least-squares editor for long sequential editing.<n>Experiments on multiple model families demonstrate stable scaling to 10K edits, outperforming strong baselines in both edit success and holistic stability.
  arXiv  Detail & Related papers  (2026-01-22T06:11:44Z)
• STABLE: Gated Continual Learning for Large Language Models [0.0]
  STABLE is a gated continual self editing framework that constrains forgetting during sequential updates.<n>Each candidate edit is evaluated against a stability budget using one of three metrics.<n>Experiments on the Qwen-2.5-7B model show that gating effectively mitigates forgetting while preserving adaptability.
  arXiv  Detail & Related papers  (2025-10-17T16:14:05Z)
• Energy-Regularized Sequential Model Editing on Hyperspheres [59.47007547581175]
  Large language models (LLMs) require constant updates to remain aligned with evolving real-world knowledge.<n> sequential editing often destabilizes representations and induces catastrophic forgetting.<n>We propose SPHERE (Sparse Projection for Hyperspherical Energy-Regularized Editing), an HE-driven regularization strategy that stabilizes neuron weight distributions.
  arXiv  Detail & Related papers  (2025-10-01T17:55:43Z)
• LyapLock: Bounded Knowledge Preservation in Sequential Large Language Model Editing [28.870053452479443]
  Current locate-then-edit approaches exhibit a progressive performance decline during sequential editing.<n>textbfLyapLock is proposed to decompose the long-term constrained programming into tractable stepwise subproblems for efficient solving.<n> 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.
  arXiv  Detail & Related papers  (2025-05-21T16:16:33Z)
• Large Continual Instruction Assistant [59.585544987096974]
  Continual Instruction Tuning (CIT) is adopted to instruct Large Models to follow human intent data by data.<n>Existing update gradient would heavily destroy the performance on previous datasets during CIT process.<n>We propose a general continual instruction tuning framework to address the challenge.
  arXiv  Detail & Related papers  (2024-10-08T11:24:59Z)
• Fine-Grained Analysis of Stability and Generalization for Stochastic Gradient Descent [55.85456985750134]
  We introduce a new stability measure called on-average model stability, for which we develop novel bounds controlled by the risks of SGD iterates. This yields generalization bounds depending on the behavior of the best model, and leads to the first-ever-known fast bounds in the low-noise setting. To our best knowledge, this gives the firstever-known stability and generalization for SGD with even non-differentiable loss functions.
  arXiv  Detail & Related papers  (2020-06-15T06:30:19Z)
• Preventing Posterior Collapse with Levenshtein Variational Autoencoder [61.30283661804425]
  We propose to replace the evidence lower bound (ELBO) with a new objective which is simple to optimize and prevents posterior collapse. We show that Levenstein VAE produces more informative latent representations than alternative approaches to preventing posterior collapse.
  arXiv  Detail & Related papers  (2020-04-30T13:27:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.