Research / Papers

Papers and research

02research papers

Browse papers by year, with research summaries, publication details, and original sources.

2026

01 records
PUBLICATION / Paper dossier
PublishedIEEE TIFS · 2026

DUAP: Disentanglement-Based Universal Adversarial Perturbations for Robust Multilingual Speech Privacy Protection

Universal multilingual adversarial perturbations for speech privacy against Whisper

Qianli MaWenjie ZhangJiahao ChenJiazhen JiaRangding WangDiqun Yan

This work addresses privacy risks from multilingual automatic speech recognition models such as Whisper through DUAP, a universal adversarial perturbation method based on language feature disentanglement. The method first separates and reconstructs language-related features in latent space, then uses gradient optimization to disrupt language identification, providing protection across languages, model sizes, and digital and physical settings.

DUAP Figure 3: language and content encoders, feature-swapping reconstruction, Whisper, and three loss functions.
Fig. 3

Language feature disentanglement (LFD): separating language and content features, then reconstructing exchanged features to generate Mel spectrograms with target-language features.

Citations and research keywords

Citation

Qianli Ma, Wenjie Zhang, Jiahao Chen, Jiazhen Jia, Rangding Wang, and Diqun Yan. DUAP: Disentanglement-Based Universal Adversarial Perturbations for Robust Multilingual Speech Privacy Protection. IEEE Transactions on Information Forensics and Security, 21:3703–3718, 2026.

DOI10.1109/TIFS.2026.3671687

Keywords

  • speech privacy
  • adversarial perturbation
  • multilingual ASR
  • Whisper
  • language disentanglement

2025

01 records
PUBLICATION / Paper dossier
PublishedAnal. Chem. · 2025

Ratiometric Determination and Discrimination of Oxicams via Dual-Excitation Carbon Dots Assisted by Machine Learning

Ratiometric detection and discrimination of oxicams with dual-excitation carbon dots and machine learning

Yihao ZhangQianli MaSineng GaoXinru LiuHaoming XingHouwen HuLinfan WangWeihao LiTing ZhangYafei HouDa Chen

This work presents a fluorescence method for detecting and distinguishing oxicam nonsteroidal anti-inflammatory drugs. It uses changes in the intensity ratio at two excitation peaks of fluorine- and nitrogen-codoped carbon dots for quantitative sensing, then combines XGBoost and convolutional neural networks for low-concentration discrimination and concentration prediction in real samples.

Citations and research keywords

Citation

Yihao Zhang, Qianli Ma, Sineng Gao, Xinru Liu, Haoming Xing, Houwen Hu, Linfan Wang, Weihao Li, Ting Zhang, Yafei Hou, and Da Chen. Ratiometric Determination and Discrimination of Oxicams via Dual-Excitation Carbon Dots Assisted by Machine Learning. Analytical Chemistry, 97(39):21428–21437, 2025.

DOI10.1021/acs.analchem.5c03226

Keywords

  • carbon dots
  • oxicams
  • fluorescence sensing
  • XGBoost
  • convolutional neural network

Image preview

Open original image