Výsledky vyhledávání

Report

Targeted Vaccine: Safety Alignment for Large Language Models against Harmful Fine-Tuning via Layer-wise Perturbation

Autor: Liu, Guozhi, Lin, Weiwei, Huang, Tiansheng, Mo, Ruichao, Mu, Qi, Shen, Li

Harmful fine-tuning attack poses a serious threat to the online fine-tuning service. Vaccine, a recent alignment-stage defense, applies uniform perturbation to all layers of embedding to make the model robust to the simulated embedding drift. However

Externí odkaz: http://arxiv.org/abs/2410.09760

Zobrazit plný text záznamu

Report

CycleNet: Enhancing Time Series Forecasting through Modeling Periodic Patterns

Autor: Lin, Shengsheng, Lin, Weiwei, Hu, Xinyi, Wu, Wentai, Mo, Ruichao, Zhong, Haocheng

The stable periodic patterns present in time series data serve as the foundation for conducting long-horizon forecasts. In this paper, we pioneer the exploration of explicitly modeling this periodicity to enhance the performance of models in long-ter

Externí odkaz: http://arxiv.org/abs/2409.18479

Zobrazit plný text záznamu

Report

SegRNN: Segment Recurrent Neural Network for Long-Term Time Series Forecasting

Autor: Lin, Shengsheng, Lin, Weiwei, Wu, Wentai, Zhao, Feiyu, Mo, Ruichao, Zhang, Haotong

RNN-based methods have faced challenges in the Long-term Time Series Forecasting (LTSF) domain when dealing with excessively long look-back windows and forecast horizons. Consequently, the dominance in this domain has shifted towards Transformer, MLP

Externí odkaz: http://arxiv.org/abs/2308.11200

Zobrazit plný text záznamu