Researchers have developed a method for detecting malicious packets that can adapt to new threats without retraining from scratch. The approach uses a self-supervised learning backbone and low-rank adaptation to efficiently update the model while preserving previously learned knowledge. This is particularly useful in the setting of few-shot class-incremental learning, where only a small number of labeled examples are available for new malware classes. Experiments across several datasets show that this method outperforms previous approaches and achieves state-of-the-art performance. --- Why it matters: This work matters to AI researchers because it addresses a key challenge in malware detection: adapting to new threats without forgetting previously learned knowledge. The proposed approach has the potential to improve the efficiency and effectiveness of malware detection systems, which is critical for ensuring network security. Source: https://arxiv.org/abs/2608.23536