Advancing AI for Low-Resource Languages

We develop intelligent technologies for low-resource languages, with a particular focus on Tibetan. Our research spans large language models, speech intelligence, machine translation, multimodal learning, and benchmark construction. By building high-quality datasets and evaluation frameworks, we aim to improve language understanding and enable trustworthy AI systems for education, healthcare, and cultural preservation.

Keywords

Large Language Models · Low-resource NLP · Tibetan AI · Speech Intelligence · Machine Translation

Representative work

------

Y. Liu, Z. Zhang, B. Ma-Bao, R. Duojie, Y. Cai, Y. Yu, X. Wang, F. Gao, C. Huang, and N. Tashi, "TMD-TTS: A Unified Tibetan Multi-Dialect Text-to-Speech Framework for Ü-Tsang, Amdo and Kham Speech Dataset Generation," in Proceedings of IEEE the 51st International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026.

------

J. Zhang, F. Gao, L. Li, Y. Yu, X. Wang, N. Tashi, and G. Luosang, "RetrieveAll: A Multilingual Named Entity Recognition Framework with Large Language Models," in Proceedings of IEEE the 51st International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026.

------

D. Tashi, B. Chen, T. Sheng, Y. Yu, X. Wang, J. Zhang, L. Yeshi, R. Dongrub, T. Tsering, and N. Tashi, "Tibetan Data Augmentation via GAN-Based Handwritten Text Generation," CAAI Transactions on Intelligence Technology, 2026.

------

F. Gao, C. Huang, Y. Liu, N. Tashi, X. Wang, T. Tsering, B. Ma-bao, R. Duojie, G. Luosang, R. Dongrub, D. Tashi, X. F. Cd, Y. Yu, and H. Wang, "TLUE: A Tibetan Language Understanding Evaluation Benchmark," in Proceedings of the 30th Conference on Empirical Methods in Natural Language Processing (EMNLP), 2025. 

→ The First Tibetan Benchmark for LLM Evaluation In The World

------

J. Zhang, Z. Zhang, L. Yeshi, D. Tashi, X. Wang, Y. Cai, Y. Yu, X. Wang, N. Tashi, and G. Luosang, "Tibetan Medical Named Entity Recognition Based on Syllable-Word-Sentence Embedding Transformer," CAAI Transactions on Intelligence Technology, 2025.

------

     ......

Intelligent Ophthalmology & Medical AI

Our research explores trustworthy AI systems for medical imaging and clinical decision support, with a particular emphasis on ophthalmology. We develop computer vision models, multimodal learning frameworks, knowledge-guided AI, and biomedical foundation models for disease diagnosis, risk assessment, and precision medicine. Our goal is to build interpretable and clinically useful AI technologies that improve healthcare outcomes.

Keywords

Medical AI · Computer Vision · Biomedical Imaging · Ophthalmology · Clinical Decision Support

Representative work

------

C. Huang, Z. Han, Y. Liu, F. Gao, J. Qiu, J. Li, W. Wei, H. Tian, Y. Zhang, Q. Dai, and Y. Yu, "CAT-Net: Image-to-Text for Medical Reports Using Adaptive Co-Attention and Triple-LSTM Module," Proceedings of the 48th IEEE Engineering in Medicine and Biology Conference (EMBC), 2026.

------

F. Ekong, Y. Yu, R. A. Patamia, K. Sarpong, C. C. Ukwuoma, X. Wang, A. R. Ukot, and J. Cai, "Masked Hybrid Attention with Laplacian Query Fusion and Tripartite Sequence Matching for Medical Image Segmentation," Neural Computing and Applications, 2025.

------

F. Ekong, Y. Yu, R. A. Patamia, K. Sarpong, C. C. Ukwuoma, A. R. Ukot, and J. Cai, "RetVes Segmentation: A Pseudo-Labeling and Feature Knowledge Distillation Optimization Technique for Retinal Vessel Channel Enhancement," Computers in Biology and Medicine, 2024.

------

Y. Lin, Q. Tang, H. Wang, C. Huang, F. Ekong, X. Wang, X. Feng, and Y. Yu, "Attention Enhanced Network with Semantic Inspector for Medical Image Report Generation," in Proceedings of IEEE the 35th International Conference on Tools with Artificial Intelligence (ICTAI), 2023.

------

C. Huang, Y. Yu, and M. Qi, "Skin Lesion Segmentation Based on Deep Learning," in Proceedings of IEEE the 20th International Conference on Communication Technology (ICCT), 2020. 

→ The Best Undergraduate Thesis in the 2020 SISE Graduation Thesis (5/710)

------

     ......

Neuromorphic Computing and Intelligent Hardware 

We investigate next-generation computing architectures based on memristive devices and neuromorphic systems. Our research focuses on in-memory computing, intelligent hardware acceleration, edge intelligence, and hardware-software co-design for efficient AI. By integrating memristor technologies with modern machine learning, we aim to develop energy-efficient intelligent computing platforms.

Keywords

Memristive Computing · Neuromorphic Computing · In-Memory AI · Intelligent Hardware · Edge Intelligence

Representative work

------

L. Gu, J. Chen, C. Huang, Z. Zhang, X. Wang, and Y. Yu, "Memristive Binary Firefly Algorithm with Orthogonal Crossing," in Proceedings of the International Symposium on Biological Neural Networks and Intelligent Optimization (BNNIO), 2026.

------

Y. Yu, Y. Liu, J. Shen, B. Ma-Bao, X. Wang, D. Tsering, T. Tsering, C. Huang, D. Tashi, R. Dongrub, R. Duojie, G. Luosang, K. Tashi, and N. Tashi, "Memristive Computing for Tibetan Spellchecker," IEEE Journal of Systems Engineering and Electronics, 2025.

------

Y. Yu, X. Wang, X. Feng, J. Shen, N. Tashi, and P. Mazumder, Memristive Computing. Springer, 2025. (Book)

------

X. Han, Y. Yu, X. Wang, X. Feng, and S. Zhong, "Stabilization of Delayed Memristive Neural Networks Driven by Mixed Deception Attacks," in Proceedings of the 2023 International Conference on Control, Automation and Diagnosis (ICCAD), 2023.

------

X. Wang, Y. Yu, J. Cai, N. Yang, K. Shi, S. Zhong, K. Adu, and N. Tashi, "Multiple Mismatched Synchronization for Coupled Memristive Neural Networks With Topology-Based Probability Impulsive Mechanism on Time Scales," IEEE Transactions on Cybernetics, 2021.

------

     ......