Ongoing Works
Dependable Retrieval-Augmented Intrusion Detection Under Retrieval Poisoning and Prompt-Injection Attacks
The 12th IEEE International Conference on Collaboration and Internet Computing (IEEE CIC 2026) Manuscript 2026
RAG-IDS addresses a critical gap in LLM-based intrusion detection: susceptibility to adversarial manipulation at the retrieval layer. By hardening the retrieval pipeline against knowledge poisoning and input-level prompt injection, RAG-IDS maintains accurate, explainable network threat detection even when the underlying knowledge base or queries are actively tampered with. Evaluated on benchmark network traffic datasets, RAG-IDS demonstrates robust detection under adversarial conditions where standard RAG-based detectors fail.
MalariAI: Automated Malaria Cell Segmentation and Classification from Blood Smear Images
Computerized Medical Imaging and Graphics (CMIG) Manuscript 2026
We propose MalariAI, a two-stage pipeline combining watershed-based cell segmentation with EfficientNet-B0 classification for automated malaria detection from blood smear images. Evaluated on NIH BBBC041 and MP-IDB datasets, achieving 75.95% cell recovery rate and 98.36% classification accuracy. Includes Grad-CAM++ visualizations and cross-dataset validation.
Publications
A Non-Invasive Cloud-Based Migration Strategy for Post-Quantum Cybersecurity in Smart HVAC Systems
IEEE Symposium on Security and Privacy (S&P) 2027 Submitted Under Review 2027
Legacy smart HVAC systems suffer from critical quantum vulnerabilities due to their reliance on classical ECDH. We propose a non-invasive, cloud-based proxy architecture that integrates NIST-standard post-quantum cryptography (ML-KEM and ML-DSA) into the ecosystem without modifying legacy hardware or firmware, providing a deployable migration pathway to quantum-safe security.
SciRet: A Compute-Aware Empirical Study of Retrieval and Reranking for Scientific RAG
Conference on Empirical Methods in Natural Language Processing (EMNLP 2026) Short Paper Submitted Under Review 2026
We introduce SciRet, a compute-aware empirical study of retrieval-augmented generation for scientific question answering over CORD-19. Rather than proposing a new model, we evaluate a fixed scientific RAG pipeline across three corpus scales: 1,034 chunks (1K papers), 5,160 chunks (5K papers), and 15,480 chunks (15K papers). The pipeline combines sentence-window chunking, BM25, BGE-M3 dense retrieval, reciprocal rank fusion, optional cross-encoder reranking, and grounded answer generation. Across these settings, hybrid retrieval is more robust than either sparse-only or dense-only retrieval in our setting, reaching Recall@10 of 1.000 at 1K and 15K. In contrast, an MS MARCO-trained cross-encoder reranker reduces precision on the scientific corpus, suggesting that domain mismatch can outweigh the benefits of stronger query-passage interaction. Generation faithfulness measured with RAGAS increases with corpus scale in our setup. Retrieval evaluation uses pseudo-relevance labels derived from the hybrid system, so we treat the results as controlled comparative evidence rather than a benchmark claim. We release code, indexes, and evaluation outputs to support replication and follow-up studies.
Accurate Prediction of Pulmonary Fibrosis Progression Using EfficientNet and Quantile Regression: A High Performing Approach
IEEE Region 10 Symposium (TENSYMP 2023) Conference 2023
Accurate prediction of pulmonary fibrosis progression is crucial for effective patient management. This study proposes an efficient deep learning framework for predicting the progression of pulmonary fibrosis using high-resolution computed tomography (HRCT) images. We leverage the EfficientNet architecture, known for its high accuracy and computational efficiency, to extract discriminative features from CT scans. To capture the uncertainty inherent in disease progression, we employ quantile regression instead of standard mean-based regression. This approach allows us to model the conditional distribution of future lung function, providing not only a point prediction but also prediction intervals that quantify the uncertainty associated with the prognosis. Our experiments on a benchmark dataset demonstrate that the proposed EfficientNet-based quantile regression model achieves state-of-the-art performance, outperforming existing methods in predicting pulmonary fibrosis progression while providing reliable uncertainty estimates.