AI Paper Atlas

Research map / Continual learning / Continual fine-tuning of LLMs

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

Xin Ma, Wei Chen, Qi Liu, Derong Xu, Zhi Zheng, Tong Xu, Enhong Chen

ICML 2026 · Continual learning › Continual fine-tuning of LLMs · approach: Continual pretraining of language models · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Lifelong Normalization to UltraEdit/RLEdit/MALMEN (ablation)UltraEdit · Llama-3-8B-Instruct, 20K edits, ZsREEfficacy: 36.8 -> 90.1Improved
add early edits before later edits (positive cumulative effect) (ablation)LN-based editors · Llama-3-8B-Instruct later segmentEfficacy: 83.9 -> 88.9 (generalization, UltraEdit)Improved
add warm-up stage to StableEdit (ablation)StableEdit · WikiBigEdit 20K, Mistral-7B-v0.3Efficacy: 76.96 -> 78.90Improved
add full whitening to StableEdit (ablation)StableEdit · WikiBigEdit 20K, Mistral-7B-v0.3Efficacy: 75.12 -> 78.90Improved
add LN to StableEdit (ablation)StableEdit · WikiBigEdit 20K, Mistral-7B-v0.3Efficacy: 39.13 -> 78.90Improved
use out-of-domain MedCF warm-up instead of in-domain warm-up (ablation)StableEdit · ZsRE/FEVER 20K, Mistral-7B-v0.3Efficacy/Gen/Spe: average absolute change of only 0.43 pointsNo clear effect
StableEdit instead of UltraEdit at 500K editsUltraEdit · WikiBigEdit 500K, GPT-J-6BGeneralization: 60.54 -> 65.04Improved
StableEdit instead of UltraEdit at standard scaleUltraEdit · UltraEditBench 20K, Llama-3-8B-InstructGeneralization: 81.28 -> 85.46Improved
StableEdit instead of UltraEdit on specificityUltraEdit · WikiBigEdit 17K, Llama-3-8B-InstructSpecificity: 48.51 -> 45.22Hurt
StableEdit instead of UltraEdit on runtimeUltraEdit · Llama-3-8B-Instructseconds per edit: 0.07 -> 0.10Hurt
remove LN from baselines/editors during update geometry analysis (ablation)StableEdit and UltraEdit · ZsRE 20K, GPT-J-6Bupdate cosine similarityImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy