Abstract: Long sequence time-series forecasting models pay excessive attention to irrelevant information and noise when predicting air temperature, leading to prediction drift and poor generalization ...

Abstract: Link prediction for directed graphs is a crucial task with diverse real-world applications. Recent advances in embedding methods and Graph Neural Networks (GNNs) have shown promising ...

ACFM ACFM: A Dynamic Spatial-Temporal Network for Traffic Prediction Pytorch ACM MM2018/A STDN Revisiting spatial-temporal similarity: A deep learning framework for traffic prediction Keras AAAI2019/A ...

A curated list of papers, tools, and resources on Multi-Token Prediction (MTP) and related techniques in Large Language Models (LLMs), Speech-Language Models (SLMs), and more. Multi-Token Prediction ...

DREAM 证明了你根本不需要任何人工标注的正负样本,只凭"一篇文档能不能帮大模型预测出正确答案"这一个信号,就能把稠密检索器训得很强;而它的关键诀窍,是把检索器算出的相似度分数注入到冻结大模型里那些"本来就在替 query 寻找相关内容"的注意力头中,让下一个词预测的损失顺着注意力这条路反向把检索器调教出来。 一、研究背景:有没有可能,不标一条数据就把检索器练强? 先说清楚这篇论文在解决什么。稠 ...

The NBA offseason often brings unexpected headlines, and the Phoenix Suns just delivered one with a strong twist of nostalgia and prediction. The team has added Michigan State legend Mateen Cleaves to ...