aiqbench analysis
This research addresses the critical challenge of classifying obfuscated fraudulent SMS messages by leveraging AI distillation techniques to transfer de-obfuscation capabilities from large teacher models to smaller, production-ready student models. The proposed ReCAST framework demonstrates that restoration-aware training significantly enhances robustness against sophisticated evasion strategies while maintaining efficiency constraints.
ReCAST: Restoration-aware Cascaded Stage-wise Training for Obfuscated SMS Risk Classification
arXiv:2609.04878v1 Announce Type: cross Abstract: 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.
Abstract: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.
Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
arXiv:2609.04878 [cs.CR]
(or arXiv:2609.04878v1 [cs.CR] for this version)
https://doi.org/10.48550/arXiv.2609.04878 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 242 words
Related coverage
Beyond Full-Model Rollback: AuroSFT for Adapter-State Multi-Task Fine-Tuning
arXiv · Aug 7, 2026
X-Restormer++: 1st Place Solution for the UG2+ CVPR 2026 All-Weather Restoration Challenge
arXiv · Jun 3, 2026
Triton-Based Optimization of Video Sparse Attention on ROCm
AMD · Jul 13, 2026
Breaking Failure Cascades: Step-Aware Reinforcement Learning for Medical Multimodal Reasoning
arXiv · Sep 25, 2026