Towards Empirical Interpretation of Internal Circuits and Properties in Grokked Transformers on Modular Polynomialsby Hiroki…
On the Generalization Capability of Temporal Graph Learning Algorithms: Theoretical Insights and a Simpler Methodby…
Spectrum Extraction and Clipping for Implicitly Linear Layersby Ali Ebrahimpour Boroojeny, Matus Telgarsky, Hari SundaramFirst…
HiGPT: Heterogeneous Graph Language Modelby Jiabin Tang, Yuhao Yang, Wei Wei, Lei Shi, Long Xia,…
InstructEdit: Instruction-based Knowledge Editing for Large Language Modelsby Ningyu Zhang, Bozhong Tian, Siyuan Cheng, Xiaozhuan…
A Duality Analysis of Kernel Ridge Regression in the Noiseless Regimeby Jihao Long, Xiaojun Peng,…
How Do Nonlinear Transformers Learn and Generalize in In-Context Learning?by Hongkang Li, Meng Wang, Songtao…
Fair Resource Allocation in Multi-Task Learningby Hao Ban, Kaiyi JiFirst submitted to arxiv on: 23…
Co-Supervised Learning: Improving Weak-to-Strong Generalization with Hierarchical Mixture of Expertsby Yuejiang Liu, Alexandre AlahiFirst submitted…
Deep Networks Always Grok and Here is Whyby Ahmed Imtiaz Humayun, Randall Balestriero, Richard BaraniukFirst…