Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2605.09664 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914564087480320 |
|---|---|
| author | Asadi, Zahra Jeon, Haeseung Han, Sohyun Kamol, Md Mahmuduzzaman Oh, Se Eun Rahman, Mohammad Saidur |
| author_facet | Asadi, Zahra Jeon, Haeseung Han, Sohyun Kamol, Md Mahmuduzzaman Oh, Se Eun Rahman, Mohammad Saidur |
| contents | As over 200 million new malware samples are identified each year, antivirus systems must continuously adapt to the evolving threat landscape. However, retraining solely on new samples leads to catastrophic forgetting and exploitable blind spots, while retraining on the entire dataset incurs substantial computational cost. We propose FreeMOCA, a memory- and compute-efficient continual learning framework for malicious code analysis that preserves prior knowledge via adaptive layer-wise interpolation between consecutive task updates, leveraging the fact that warm-started task optima are connected by low-loss paths in parameter space.
We evaluate FreeMOCA in both class-incremental (Class-IL) and domain-incremental (Domain-IL) settings on large-scale Windows (EMBER) and Android (AZ) malware benchmarks. FreeMOCA achieves substantial gains in Class-IL, outperforming 11 baselines on both EMBER and AZ benchmarks. It also significantly reduces forgetting, achieving the best retention across baselines, and improving accuracy by up to 42% and 37% on EMBER and AZ, respectively. These results demonstrate that warm-started interpolation in parameter space provides a scalable and effective alternative to replay for continual malware detection.
Code is available at: https://github.com/IQSeC-Lab/FreeMOCA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_09664 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | FreeMOCA: Memory-Free Continual Learning for Malicious Code Analysis Asadi, Zahra Jeon, Haeseung Han, Sohyun Kamol, Md Mahmuduzzaman Oh, Se Eun Rahman, Mohammad Saidur Cryptography and Security Machine Learning As over 200 million new malware samples are identified each year, antivirus systems must continuously adapt to the evolving threat landscape. However, retraining solely on new samples leads to catastrophic forgetting and exploitable blind spots, while retraining on the entire dataset incurs substantial computational cost. We propose FreeMOCA, a memory- and compute-efficient continual learning framework for malicious code analysis that preserves prior knowledge via adaptive layer-wise interpolation between consecutive task updates, leveraging the fact that warm-started task optima are connected by low-loss paths in parameter space. We evaluate FreeMOCA in both class-incremental (Class-IL) and domain-incremental (Domain-IL) settings on large-scale Windows (EMBER) and Android (AZ) malware benchmarks. FreeMOCA achieves substantial gains in Class-IL, outperforming 11 baselines on both EMBER and AZ benchmarks. It also significantly reduces forgetting, achieving the best retention across baselines, and improving accuracy by up to 42% and 37% on EMBER and AZ, respectively. These results demonstrate that warm-started interpolation in parameter space provides a scalable and effective alternative to replay for continual malware detection. Code is available at: https://github.com/IQSeC-Lab/FreeMOCA. |
| title | FreeMOCA: Memory-Free Continual Learning for Malicious Code Analysis |
| topic | Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2605.09664 |