Unified EEG Representation Learning
Transferable representations across heterogeneous EEG tasks, temporal scales, neural patterns, and label spaces.
HELLO, I’M
Haiyang Lu
I am a master's student in Computer Technology at Tongji University. I work on EEG foundation models, LLM post-training, distributed training, and AI agents—with an interest in turning learning algorithms into useful intelligent systems.
EEG Foundation ModelsLLM Post-trainingAI AgentsDistributed Training
10,000+
EEG pretraining hours
6
EEG task families
200K
Multimodal EEG pairs
2
Open-source agent projects
RESEARCH INTERESTS
What I Work On
Efficient Adaptation & Deployment
Contrastive adaptation, Mixture-of-Experts, quantization, pruning, and mixed-precision training.
LLM Post-training & Agents
Reinforcement learning, repository-level code agents, retrieval, tool use, and reliable agent workflows.
FEATURED PUBLICATION
Latest Research
ACM MM 2026 · ACCEPTEDFirst Author
UniNeuro: A Unified EEG Foundation Model for Heterogeneous EEG Task Learning
Haiyang Lu, Lianghua He, Hongzhou Chen, Xiao Chen, Wenqi Zhang
A unified foundation model for heterogeneous EEG tasks. Task-conditional MoE disentangles shared and task-specific representations, while hierarchical prototype-guided contrastive learning improves downstream adaptation.
SELECTED WORK
Projects
BACKGROUND
Education
Tongji University
M.Eng. in Computer Technology
2024–2027 (expected)
Sichuan University
B.Eng. in Software Engineering
2020–2024
WRITING
Latest Notes
The first post on my personal website—and a new beginning.
Let’s connect
I am always happy to discuss EEG foundation models, post-training, and agent systems.