You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Research Interests: Self-Supervised Pretraining, Cross-Modal Representation Alignment and Fusion, Robust Representation Learning, and Brain-Computer Interface
A visual grounding system for RefCOCO that freezes GroundingDINO candidates and trains a lightweight multimodal fusion reranker for region-text alignment and verifiable candidate ranking.
This repository implements a robust, multi-stage self-supervised learning framework designed for multi-channel time-series physiological signals (e.g., EEG). Addressing the common challenges of lim…