Search Results for author:
Found 2 papers, 2 papers with code
SeDR: Segment Representation Learning for Long Documents Dense Retrieval
In SeDR, Segment-Interaction Transformer is proposed to encode long documents into document-aware and segment-sensitive representations, while it holds the complexity of splitting-and-pooling and outperforms other segment-interaction patterns on DR.
Personalized Cross-Silo Federated Learning on Non-IID Data
Non-IID data present a tough challenge for federated learning.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
