Summary of Loki: Low-rank Keys For Efficient Sparse Attention, by Prajwal Singhania et al.
Loki: Low-rank Keys for Efficient Sparse Attentionby Prajwal Singhania, Siddharth Singh, Shwai He, Soheil Feizi,…
Loki: Low-rank Keys for Efficient Sparse Attentionby Prajwal Singhania, Siddharth Singh, Shwai He, Soheil Feizi,…
Learning to grok: Emergence of in-context learning and skill composition in modular arithmetic tasksby Tianyu…
Progressive Confident Masking Attention Network for Audio-Visual Segmentationby Yuxuan Wang, Jinchao Zhu, Feng Dong, Shuyue…
What Improves the Generalization of Graph Transformers? A Theoretical Dive into the Self-attention and Positional…
FFNet: MetaMixer-based Efficient Convolutional Mixer Designby Seokju Yun, Dongheon Lee, Youngmin RoFirst submitted to arxiv…
Iteration Head: A Mechanistic Study of Chain-of-Thoughtby Vivien Cabannes, Charles Arnal, Wassim Bouaziz, Alice Yang,…
CAFO: Feature-Centric Explanation on Time Series Classificationby Jaeho Kim, Seok-Ju Hahn, Yoontae Hwang, Junghye Lee,…
A Global Geometric Analysis of Maximal Coding Rate Reductionby Peng Wang, Huikang Liu, Druv Pai,…
Position: Cracking the Code of Cascading Disparity Towards Marginalized Communitiesby Golnoosh Farnadi, Mohammad Havaei, Negar…
DiffUHaul: A Training-Free Method for Object Dragging in Imagesby Omri Avrahami, Rinon Gal, Gal Chechik,…