What Improves the Generalization of Graph Transformers? A Theoretical Dive into the Self-attention and Positional…
Position Coupling: Improving Length Generalization of Arithmetic Transformers Using Task Structureby Hanseul Cho, Jaeyoung Cha,…
CycleFormer : TSP Solver Based on Language Modelingby Jieun Yook, Junpyo Seo, Joon Huh, Han…
PoPE: Legendre Orthogonal Polynomials Based Position Encoding for Large Language Modelsby Arpit AggarwalFirst submitted to…
Predictive Modeling with Temporal Graphical Representation on Electronic Health Recordsby Jiayuan Chen, Changchang Yin, Yuanlong…
CKGConv: General Graph Convolution with Continuous Kernelsby Liheng Ma, Soumyasundar Pal, Yitian Zhang, Jiaming Zhou,…
Cost-Efficient Large Language Model Serving for Multi-turn Conversations with CachedAttentionby Bin Gao, Zhuomin He, Puru…
MEP: Multiple Kernel Learning Enhancing Relative Positional Encoding Length Extrapolationby Weiguo GaoFirst submitted to arxiv…
Pretraining Codomain Attention Neural Operators for Solving Multiphysics PDEsby Md Ashiqur Rahman, Robert Joseph George,…
Found in the Middle: How Language Models Use Long Contexts Better via Plug-and-Play Positional Encodingby…