Qiang Zhang

The paper proposes CoDe-R, a two-stage framework that significantly improves the accuracy and re-executability of decompiled code generated by LLMs, achieving a new SOTA in the lightweight regime.

The paper proposes HTell, a fast and lightweight data-free backdoor detector that analyzes the abnormal response concentration of backdoored models on the target class using random latent probes applied directly to the prediction head.

The paper proposes DFBScanner, a lightweight static parameter inspection framework that detects backdoor attacks by analyzing anomalous parameter updates in the final classification layer, achieving fast and generalizable detection.

The paper proposes DAG-MoE, a novel sparse Mixture-of-Experts framework that replaces standard weighted-sum aggregation with structural aggregation to enhance model performance and enable multi-step reasoning.

The paper proposes PC-MambaSDE, a physically-constrained continuous-time framework that accurately predicts Remaining Useful Life (RUL) despite irregular sensor observations and ensures physically plausible degradation trajectories.

The paper proposes Value-order Decomposition (VOD) to address the challenge of cross-domain generalization in industrial anomaly detection by disentangling and suppressing object-category-, defect-type-, and domain-specific information.

The paper introduces GigaSpeechBench, a comprehensive multilingual and multidimensional ASR & AST benchmark with 680 hours of human-annotated speech, featuring 12 low-resource languages, 6 Chinese dialects, 6 English accents, dense terminology, older adult and child speech, and human-annotated translations.