People › Wenqiang Luo
Wenqiang Luo

Wenqiang Luo

羅文强 PhD Student Google Scholar ↗CityU Scholars ↗

About

Wenqiang Luo is a PhD Student in the ASI Lab. of Department of Computer Science, City University of Hong Kong.

Wenqiang Luo works on automated program repair and software log analysis, with emphasis on privacy-preserving and low-resource settings. His federated learning study shows privacy-preserving federated fine-tuning on proprietary code significantly enhances program repair, improving Top@10 by up to 16.67% and Pass@10 by 18.44% while approaching centralized-learning quality, with a negligible impact from code heterogeneity. He proposes LogMeta, a MAML-based few-shot framework with a hybrid RoBERTa/Bi-LSTM model that improves F1 by up to 28.3% in low-resource log anomaly detection. He also introduces LANTERN, a cross-language multi-agent repair approach that improves Rust Pass@10 by 22.09% on xCodeEval.

Research Interests

Publications (5)