Zixi Wang
MS student in Software Engineering
University of Electronic Science and Technology of China
I am interested in machine learning, gradual domain adaptation, and robust visual perception. My recent work explores how to transfer knowledge across continuously changing domains with stable training strategies.
Selected Publications
-
Self-Training with Dynamic Weighting for Robust Gradual Domain Adaptation
TL;DR: Studies robust gradual domain adaptation with a self-training strategy and dynamic weighting under continuously shifting domains.
The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025
-
Multi-Modal Gradual Domain Osmosis: Stepwise Dynamic Learning with Batch Matching for Gradual Domain Adaptation
TL;DR: Introduces a multimodal gradual adaptation framework with stepwise dynamic learning and batch matching.
The 33rd ACM International Conference on Multimedia (ACMMM), 2025
Projects
-
AiTeachMe
A personal AI learning workspace for material parsing, study planning, knowledge maps, companion Q&A, and learner profiling.
-
Idea2PaperLab
An agent workspace for idea refinement, literature expansion, experiment execution, result analysis, and LaTeX drafting.
-
FluxHive
An edge-cloud GPU scheduling platform with memory-aware dispatch, streaming logs, RBAC, auditing, and realtime monitoring.