NeuSemSlice: Towards Effective DNN Model Maintenance via Neuron-Level Semantic Slicing
Open Access
- 11 December 2025
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Software Engineering and Methodology
- Vol. 35 (1) , 1-27
Abstract
Deep Neural Networks (DNNs), extensively applied across diverse disciplines, are characterized by their integrated and monolithic architectures, setting them apart from conventional software systems. This architectural difference introduces particular challenges to maintenance tasks, such as model restructure (e.g., model compression), re-adaptation (e.g., fitting new samples), and incremental development (e.g., continual knowledge accumulation). Prior research addresses these challenges by identifying task-critical neuron layers and dividing neural networks into semantically similar sequential modules. However, such layer-level approaches fail to precisely identify and manipulate neuron-level semantic components, restricting their applicability to finer-grained model maintenance tasks. In this work, we implement NeuSemSlice, a novel framework that introduces the semantic slicing technique to effectively identify critical neuron-level semantic components in DNN models for semantic-aware model maintenance tasks. Specifically, semantic slicing identifies, categorizes, and merges critical neurons across different categories and layers according to their semantic similarity, enabling their flexibility and effectiveness in the subsequent tasks. For semantic-aware model maintenance tasks, we provide a series of novel strategies based on semantic slicing to enhance NeuSemSlice. They include semantic components (i.e., critical neurons) preservation for model restructure, critical neuron tuning for model re-adaptation, and non-critical neuron training for model incremental development. A thorough evaluation has demonstrated that NeuSemSlice significantly outperforms baselines in all three tasks.Keywords
All Related Versions
Funding Information
- National NSF of China (62302176, 62072046, 62302181)
- Key R&D Program of Hubei Province (2023BAB017, 2023BAB079)
- Knowledge Innovation Program of Wuhan-Basic Research (2022010801010083)
This publication has 39 references indexed in Scilit:
- MalWuKong: Towards Fast, Accurate, and Multilingual Detection of Malicious Code Poisoning in OSS Supply Chains2023
- An application of deep reinforcement learning to algorithmic tradingExpert Systems with Applications, 2021
- Understanding adversarial robustness via critical attacking routeInformation Sciences, 2021
- Importance Estimation for Neural Network Pruning2019
- Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks2018
- Deep neural architectures for prediction in healthcareComplex & Intelligent Systems, 2017
- Managing software evolution through semantic history slicing2017
- Deep Residual Learning for Image Recognition2016
- ImageNet: A large-scale hierarchical image database2009
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998