Complex Pattern
Complex pattern analysis focuses on identifying, understanding, and utilizing recurring structures within diverse data types, ranging from human mobility and network communications to images and time series. Current research emphasizes developing novel algorithms and model architectures, such as those based on tensor factorization, recurrent neural networks, and generative models, to efficiently extract and interpret these patterns, often incorporating temporal dynamics and handling incomplete or noisy data. This field is crucial for advancing various scientific disciplines and practical applications, including anomaly detection, predictive modeling, and improved decision-making in areas like healthcare, finance, and autonomous systems.
Papers
Explainable AI for Multivariate Time Series Pattern Exploration: Latent Space Visual Analytics with Time Fusion Transformer and Variational Autoencoders in Power Grid Event Diagnosis
Haowen Xu, Ali Boyaci, Jianming Lian, Aaron Wilson
JailPO: A Novel Black-box Jailbreak Framework via Preference Optimization against Aligned LLMs
Hongyi Li, Jiawei Ye, Jie Wu, Tianjie Yan, Chu Wang, Zhixin Li