Unified Framework
Unified frameworks in machine learning aim to consolidate diverse approaches to a specific problem into a single, coherent architecture, improving efficiency and facilitating comparative analysis. Current research focuses on developing such frameworks for various tasks, including recommendation systems, video understanding, and natural language processing, often leveraging transformer models, diffusion models, and recurrent neural networks. These unified approaches enhance model performance, enable more robust comparisons between methods, and offer improved interpretability and controllability, ultimately advancing both theoretical understanding and practical applications across numerous domains.
Papers
PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback
Souradip Chakraborty, Amrit Singh Bedi, Alec Koppel, Dinesh Manocha, Huazheng Wang, Mengdi Wang, Furong Huang
Neural Collapse Terminus: A Unified Solution for Class Incremental Learning and Its Variants
Yibo Yang, Haobo Yuan, Xiangtai Li, Jianlong Wu, Lefei Zhang, Zhouchen Lin, Philip Torr, Dacheng Tao, Bernard Ghanem