In the paradigm of multi-task learning, mul- tiple related prediction tasks are learned jointly, sharing information across the tasks. We propose a framework for multi-task learn- ing that enables one to selectively share the information across the tasks. We assume that each task parameter vector is a linear combi- nation of a finite number of underlying basis tasks. The coefficients of the linear combina- tion are sparse in nature and the overlap in the sparsity patterns of two tasks controls the amount of sharing across these. Our model is based on on the assumption that task pa- rameters within a group lie in a low dimen- sional subspace but allows the tasks in differ- ent groups to overlap with each other in one or more bases. Experimental results on four datasets show that our approach outperforms competing methods.
A new model for natural groupings in high-dimensional data
Eindhoven Institute of Technology(埃因霍温理工大学);Elmore Family School of Electrical and Computer Engineering, Purdue University(普渡大学埃尔莫尔家族电气与计算机工程学院);Department of Mathematics, Purdue University(普渡大学数学系)
Multiple Fairness and Cardinality constraints for Students-Topics Grouping Problem
L3S Research Center(L3S研究中心);Leibniz University Hannover(汉诺威莱布尼茨大学);Institute for Didactics of Mathematics and Physics(数学与物理教学法研究所);Institute of Computer Science(计算机科学研究所);Free University Berlin(柏林自由大学)
MINT: Multimodal Instruction Tuning with Multimodal Interaction Grouping
University of California, San Diego(加州大学圣地亚哥分校);Johns Hopkins University(约翰霍普金斯大学);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Massachusetts Institute of Technology(麻省理工学院)
本文提出 MINT 任务分组策略,按多模态交互类型组织指令微调任务,以学习可迁移技能并减少任务干扰,在泛化与专业化间取得更优平衡。