ReCAST: Restoration-aware Cascaded Stage-wise Training for Obfuscated SMS Risk Classification
ReCAST distills a large teacher model's de-obfuscation ability into smaller models for robust classification of obfuscated Chinese SMS fraud messages.
The paper proposes ReCAST, a restoration-aware cascaded stage-wise training framework for classifying obfuscated Chinese SMS messages. It distills a large teacher model's de-obfuscation capability into a smaller deployable student by supervising obfuscated span detection, obfuscation type prediction, and text restoration, then uses the student for risk classification. On an internally constructed real-world Chinese SMS benchmark, ReCAST substantially outperforms directly trained baselines under obfuscation, targeting production latency and throughput constraints.
- Targets obfuscation strategies used to evade SMS fraud classifiers
- Distills teacher de-obfuscation ability into a deployable student model
- Supervises span detection, obfuscation type prediction, and text restoration
- Outperforms baselines on a real-world Chinese SMS benchmark
Full article144 words · extracted from arxiv.org · click to collapse
Fraudulent messages sent via Short Message Service (SMS) are increasingly obfuscated to evade cost-conscious classifiers in production systems. In Chinese SMS, attackers can exploit a wide range of carefully crafted obfuscation strategies to hide risk-bearing phrases while preserving human readability, making direct classification brittle under real-world latency and throughput constraints. We propose ReCAST, a Restoration-aware Cascaded Stage-wise Training framework for robust obfuscated Chinese SMS classification. ReCAST distills a large teacher model's de-obfuscation ability into a smaller deployable student model by supervising obfuscated span detection, obfuscation type prediction, and text restoration, and then uses the restoration-aware student for downstream risk classification. Experiments on an internally constructed real-world Chinese SMS benchmark show that ReCAST substantially improves classification performance over directly trained baselines under obfuscation. The results suggest that restoration-aware distillation offers a practical path toward robust SMS risk classification with smaller deployable models under production-oriented constraints.
Text extracted automatically; images, tables and formatting may be missing. Original: https://arxiv.org/abs/2609.04878