Výsledky vyhledávání

Report

Infinite-LLM: Efficient LLM Service for Long Context with DistAttention and Distributed KVCache

Autor: Lin, Bin, Zhang, Chen, Peng, Tao, Zhao, Hanyu, Xiao, Wencong, Sun, Minmin, Liu, Anmin, Zhang, Zhipeng, Li, Lanbo, Qiu, Xiafei, Li, Shen, Ji, Zhigang, Xie, Tao, Li, Yong, Lin, Wei

Large Language Models (LLMs) demonstrate substantial potential across a diverse array of domains via request serving. However, as trends continue to push for expanding context sizes, the autoregressive nature of LLMs results in highly dynamic behavio

Externí odkaz: http://arxiv.org/abs/2401.02669

Zobrazit plný text záznamu

Report

Safety and Performance, Why Not Both? Bi-Objective Optimized Model Compression against Heterogeneous Attacks Toward AI Software Deployment

Autor: Zhu, Jie, Wang, Leye, Han, Xiao, Liu, Anmin, Xie, Tao

The size of deep learning models in artificial intelligence (AI) software is increasing rapidly, hindering the large-scale deployment on resource-restricted devices (e.g., smartphones). To mitigate this issue, AI software compression plays a crucial

Externí odkaz: http://arxiv.org/abs/2401.00996

Zobrazit plný text záznamu

Report

Sampling-based Fast Gradient Rescaling Method for Highly Transferable Adversarial Attacks

Autor: Han, Xu, Liu, Anmin, Yao, Chenxuan, Fan, Yanbo, He, Kun

Deep neural networks are known to be vulnerable to adversarial examples crafted by adding human-imperceptible perturbations to the benign input. After achieving nearly 100% attack success rates in white-box setting, more focus is shifted to black-box

Externí odkaz: http://arxiv.org/abs/2307.02828

Zobrazit plný text záznamu

Report

Sampling-based Fast Gradient Rescaling Method for Highly Transferable Adversarial Attacks

Autor: Han, Xu, Liu, Anmin, Xiong, Yifeng, Fan, Yanbo, He, Kun

Deep neural networks have shown to be very vulnerable to adversarial examples crafted by adding human-imperceptible perturbations to benign inputs. After achieving impressive attack success rates in the white-box setting, more focus is shifted to bla

Externí odkaz: http://arxiv.org/abs/2204.02887

Zobrazit plný text záznamu