Tianyang Wang, Ph.D.

Assistant Professor
Department of Computer Science
The University of Alabama at Birmingham
Office: University Hall 4157
Phone: 205-934-0650
Email: tw2@uab.edu or toseattle@siu.edu

About me

I joined the University of Alabama at Birmingham in August 2023, serving the department of computer science as a tenure-track assistant professor. I earned my Ph.D. degree in computer science from Southern Illinois University, and my master (computer science) and bachelor (software engineering) degrees from Jilin University, China. I was a visiting researcher at Baidu Research in 2019. I have been working as a remote researcher (unpaid) in the XuLab at Carnegie Mellon University (CMU) since 2020.

Ph.D. RA Positions: I am actively hiring Ph.D. students to join my group as research assistants. All the positions will be fully funded. The students will have a chance to be recommended to do a research internship at CMU, Oak Ridge National Laboratory, or Amazon during the Ph.D. study. Interested students are encouraged to email your CV to: tw2@uab.edu .


Research

My research interests include AI, machine (deep) learning, computer vision, NLP, and broad data science. I am especially interested in leveraging AI and machine learning for interdisciplinary research, such as biomedical informatics. Our current works focus on 1. multi-modal large language models and their parameter-efficient adaptation, 2. multi-modal video generation, 3. deep active learning, 4. deep transfer learning, and 5. their applications in various domains (biomedical research, intelligent transportation system, natural disaster analysis, etc). Please refer to the publications for details.

Publications (60+)
I have published over 60 papers in well-known journals and conferences, including the top venues (highlighted): TNNLS, NeurIPS, ICCV, AAAI, IJCAI, ACM MM, MICCAI, EMNLP, and COLM.

[AAAI] CAD-VAE: Leveraging Correlation-Aware Latents for Comprehensive Fair Disentanglement, Link, 2026.

[WACV] RoadBench: A Vision-Language Foundation Model and Benchmark for Road Damage Understanding, Link, Round 1 Acceptance (85/1329 ≈ 6.4%), 2026.

[WACV] PROBE: Self-Supervised Visual Prompting for Cross-Domain Road Damage Detection, Link available soon, 2026.

[WACV] 4D Multimodal Co-attention Fusion Network with Latent Contrastive Alignment for Alzheimer's Diagnosis, Link, 2026.

[NeurIPS] MoRE-Brain: Routed Mixture of Experts for Interpretable and Generalizable Cross-Subject fMRI Visual Decoding, Link, 2025.

[ACM MM] Visual Instance-aware Prompt Tuning, Link, 2025.

[ICCV] MagicID: Hybrid Preference Optimization for ID-Consistent and Dynamic-Preserved Video Customization, Link, 2025.

[COLM] M²IV: Towards Efficient and Fine-grained Multimodal In-Context Learning in Large Vision-Language Models, Link, 2025.

[EMNLP Findings] Sensitivity-LoRA : Low-Load Sensitivity-Based Fine-Tuning for Large Language Models, Link, 2025.

[IJCAI] Faster Annotation for Elevation-Guided Flood Extent Mapping by Consistency-Enhanced Active Learning, Link, 2025.

[ICASSP] TD-RD: A Top-Down Benchmark with Real-Time Framework for Road Damage Detection, Link, 2025.

[BMVC] Semi-MoE: Mixture-of-Experts meets Semi-Supervised Histopathology Segmentation, Link, 2025.

[CVPRW] Multimodal Generalized Category Discovery, Link, 2025.

[ICCVW] Describe Anything Model for Visual Question Answering on Text-rich Images, Link, 2025.

[ICMLW] Describe Anything in Medical Images, Link, 2025.

[CVPRW] Visual Variational Autoencoder Prompt Tuning, Link, 2025.

[MDPI Remote Sensing] Visual Prompt Learning of Foundation Models for Post-Disaster Damage, Link, 2025.

[MIPR] Mitigating Image Captioning Hallucinations in Vision-Language Models, Link, 2025.

[arXiv] Prompt-based Adaptation in Large-scale Vision Models: A Survey, Link, 2025.

[arXiv] CTR-LoRA: Curvature-Aware and Trust-Region Guided Low-Rank Adaptation for Large Language Models, Link, 2025.

[arXiv] Towards Foundation Models for Cryo-ET Subtomogram Analysis, Link, 2025.

[arXiv] CryoCCD: Conditional Cycle-consistent Diffusion with Biophysical Modeling for Cryo-EM Synthesis, Link, 2025.

[arXiv] AutoMiSeg: Automatic Medical Image Segmentation via Test-Time Adaptation of Foundation Models, Link, 2025.

[arXiv] Beyond Editing Pairs: Fine-Grained Instructional Image Editing via Multi-Scale Learnable Regions, Link, 2025.

[arXiv] SaSi: A Self-augmented and Self-interpreted Deep Learning Approach for Few-shot Cryo-ET Particle Detection, Link, 2025.

[arXiv] FOCUS: Fused Observation of Channels for Unveiling Spectra, Link, 2025.

[MICCAI] CryoSAM: Training-free CryoET Tomogram Segmentation with Foundation Models, Link, 2024.

[AAAI] Deep Active Learning with Noise Stability, Link, 2024.

[ICONIP] HGTDP-DTA: Hybrid Graph-Transformer with Dynamic Prompt for Drug-Target Binding Affinity Prediction, Link, 2024.

[ICTAI] Multi-dimension Transformer with Attention-based Filtering for Medical Image Segmentation, Link, 2024.

[IJCIS] MocFormer: A Two-Stage Pre-training-Driven Transformer for Drug–Target Interactions Prediction, Link, 2024.

[arXiv] Enhancing Weakly Supervised 3D Medical Image Segmentation through Probabilistic-aware Learning, Link, 2024.

[arXiv] Deep Active Learning with Manifold-preserving Trajectory Sampling, Link, 2024.

[arXiv] DenseMP: Unsupervised Dense Pre-training for Few-shot Medical Image Segmentation, Link, 2024.

[arXiv] Cycle-YOLO: An Efficient and Robust Framework for Pavement Damage Detection, Link, 2024.

[arXiv] Lipschitz-Driven Noise Robustness in VQ-AE for High-Frequency Texture Repair in ID-Specific Talking Heads, Link, 2024.

[arXiv] Uncertainty-Aware Adapter: Adapting Segment Anything Model (SAM) for Ambiguous Medical Image Segmentation, Link, 2024.

[ICCV] Towards Inadequately Pre-trained Models in Transfer Learning, Link, 2023.

[ECML-PKDD] Overcoming Catastrophic Forgetting for Fine-tuning Pre-trained GANs, Link, 2023.

[ICASSP] Improving BERT Fine-tuning via Stabilizing Cross-layer Mutual Information, Link, 2023.

[TNNLS] Temporal Output Discrepancy for Loss Estimation-based Active Learning, Link, 2022.

[AAAI] Boosting Active Learning via Improving Test Performance, Link, 2022.

[ICIP] Deep Active Learning for Cryo-Electron Tomography Classification, Link, 2022.

[ICASSP] Parameter-Free Style Projection for Arbitrary Style Transfer, Link, 2022.

[IEEE Access] Deep-Precognitive Diagnosis: Preventing Future Pandemics by Novel Disease Detection with Biologically-inspired Conv-Fuzzy Network, Link, 2022.

[IEEE Access] Prolificacy Assessment of Spermatozoan via State-of-the-Art Deep Learning Frameworks, Link, 2022.

[ICCV] Semi-Supervised Active learning with temporal Output Discrepancy, Link, 2021.

[VISAPP] Single Image Super-resolution using Vectorization and Texture Synthesis, Link, 2021.

[CVIU] A Comparison of Methods for 3D Scene Shape Retrieval, Link, 2020.

[ICTAI] I2S2: Image-to-Scene Sketch Translation Using Conditional Input and Adversarial Networks, Link, 2020.

[Molecular Carcinogenesis] Protective Role of Histone Deacetylase 4 from Ultraviolet Radiation-Induced DNA Lesions, Link, 2020.

[MIPR] Semantic Tree-Based 3D Scene Model Recognition, Link, 2020.

[WACV] Instance-based Deep Transfer Learning, Link, 2019.

[ICIP] Temporal Interframe Pattern Analysis for Static and Dynamic Hand Gesture Recognition, Link, 2019.

[ICTAI] Rethink Gaussian Denoising Prior for Real-World Image Denoising, Link, 2019.

[3DOR] SHREC’19 Track: Extended 2D Scene Image-Based 3D Scene Retrieval, Link, 2019.

[3DOR] SHREC’19 Track: Extended 2D Scene Sketch-Based 3D Scene Retrieval, Link, 2019.

[ICTAI] Data Dropout: Optimizing Training Data for Convolutional Neural Networks, Link, 2018.

[ICTAI] Dilated Deep Residual Network for Image Denoising, Link, 2017.

[ICONIP] An ELU Network with Total Variation for Image Denoising, Link, 2017.

[ICDIP] A Visual Perceptual Descriptor with Depth Feature for Image Retrieval, Link, 2017.

[arXiv] Imbalanced Malware Images Classification: a CNN based Approach, Link, 2017.

Software

T. Wang, X. Li, P. Yang, G. Hu, X. Zeng, S. Huang, C. Xu, and M. Xu. AL-GradNorm.

S. Huang, T. Wang, H. Xiong, J. Huan, and D. Dou. TOD.

S. Huang, H. Xiong, T. Wang, B. Wen, Q. Wang, Z. Chen, J. Huan, and D. Dou. Stylepro_Artistic.


Services

I regularly serve academic journals and conferences as a reviewer. The venues are listed as follows.
Journal
  • IEEE Transactions on Multimedia (TMM)
  • Elsevier Journal of Pattern Recognition (PR)
  • Elsevier Journal of Computer Vision and Image Understanding (CVIU)
  • Elsevier Journal of Neurocomputing
  • Elsevier Journal of Knowledge-Based Systems (KBS)
  • Elsevier Journal of Computer-Aided Design (CAD)
  • International Journal on Artificial Intelligence Tools (IJAIT)
Conference
  • Annual Conference on Neural Information Processing Systems (NeurIPS)
  • IEEE International Conference on Computer Vision (ICCV)
  • IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • The European Conference on Computer Vision (ECCV)
  • IEEE Winter Conference on Applications of Computer Vision (WACV)
  • IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • The European Conference on Artificial Intelligence (ECAI)
  • The British Machine Vision Conference (BMVC)

Ph.D. students

Xi Xiao (Personal Page), Spring 2024 - Present.    Xi interns at Oak Ridge National Laboratory during Summer 2025.

Teaching
At the University of Alabama at Birmingham, I teach the following courses.
  • CS 420/520 Software Engineering, Fall.
  • CS 667/767 Machine Learning, Spring.
  • CS 665/765 Deep Learning, Fall.
Prior to joining the University of Alabama at Birmingham, I taught the following courses at the Austin Peay State University.
  • CSCI 1010 Introduction to Programming I
  • CSCI 2700 Data Communications and Networking
  • CSCI 3000 Data Modeling
  • CSCI 3400 Computer Organization I
  • CSCI 4450 Introduction to Artificial Intelligence
  • CSCI 5010 Database Management Concepts
  • CSCI 5015 Data Science in Python
  • CSCI 5040 Big Data Modeling and Management