Learning from Yesterday's Error: An Efficient Online Learning Method for Traffic Demand Prediction
arxiv
MEG-XL: Data-Efficient Brain-to-Text via Long-Context Pre-Training
arxiv
Cut Your Losses! Learning to Prune Paths Early for Efficient Parallel Reasoning
arxiv
The Key to State Reduction in Linear Attention: A Rank-based Perspective
arxiv
Enhancing Multivariate Time Series Forecasting with Global Temporal Retrieval
arxiv