Multi Layer
Multi-layer architectures are a central theme in contemporary machine learning, aiming to improve the efficiency and accuracy of various models by strategically organizing computational units into multiple layers. Current research focuses on optimizing these architectures, exploring alternatives to traditional multilayer perceptrons (MLPs) such as Kolmogorov-Arnold Networks (KANs) and Fourier Analysis Networks (FANs), and investigating techniques like layer distillation and frequency shifting for improved performance and reduced computational cost. These advancements have significant implications for diverse applications, including music generation, image processing, natural language processing, and scientific computing, by enabling faster, more accurate, and more efficient models.
Papers
A Layered Architecture for Developing and Enhancing Capabilities in Large Language Model-based Software Systems
Dawen Zhang, Xiwei Xu, Chen Wang, Zhenchang Xing, Robert Mao
Error-Feedback Model for Output Correction in Bilateral Control-Based Imitation Learning
Hiroshi Sato, Masashi Konosu, Sho Sakaino, Toshiaki Tsuji
Layer Importance and Hallucination Analysis in Large Language Models via Enhanced Activation Variance-Sparsity
Zichen Song, Sitan Huang, Yuxin Wu, Zhongfeng Kang
Seeing Clearly by Layer Two: Enhancing Attention Heads to Alleviate Hallucination in LVLMs
Xiaofeng Zhang, Yihao Quan, Chaochen Gu, Chen Shen, Xiaosong Yuan, Shaotian Yan, Hao Cheng, Kaijie Wu, Jieping Ye
LayerDAG: A Layerwise Autoregressive Diffusion Model for Directed Acyclic Graph Generation
Mufei Li, Viraj Shitole, Eli Chien, Changhai Man, Zhaodong Wang, Srinivas Sridharan, Ying Zhang, Tushar Krishna, Pan Li
Multi-modal biometric authentication: Leveraging shared layer architectures for enhanced security
Vatchala S, Yogesh C, Yeshwanth Govindarajan, Krithik Raja M, Vishal Pramav Amirtha Ganesan, Aashish Vinod A, Dharun Ramesh