DenseFormer: Enhancing Information Flow in Transformers via Depth Weighted Averagingby Matteo Pagliardini, Amirkeivan Mohtashami, Francois…
CompeteSMoE – Effective Training of Sparse Mixture of Experts via Competitionby Quang Pham, Giang Do,…
Timer: Generative Pre-trained Transformers Are Large Time Series Modelsby Yong Liu, Haoran Zhang, Chenyu Li,…
Learning Mutual Excitation for Hand-to-Hand and Human-to-Human Interaction Recognitionby Mengyuan Liu, Chen Chen, Songtao Wu,…
INViT: A Generalizable Routing Problem Solver with Invariant Nested View Transformerby Han Fang, Zhihao Song,…
Minusformer: Improving Time Series Forecasting by Progressively Learning Residualsby Daojun Liang, Haixia Zhang, Dongfeng Yuan,…
Arithmetic Feature Interaction Is Necessary for Deep Tabular Learningby Yi Cheng, Renjun Hu, Haochao Ying,…
Transolver: A Fast Transformer Solver for PDEs on General Geometriesby Haixu Wu, Huakun Luo, Haowen…
Frequency Explains the Inverse Correlation of Large Language Models’ Size, Training Data Amount, and Surprisal’s…
Enhancing Cross-domain Link Prediction via Evolution Process Modelingby Xuanwen Huang, Wei Chow, Yize Zhu, Yang…