Výsledky vyhledávání - "Zhang Jianhai"

Report

Robust Diffusion Models for Adversarial Purification

Autor: Lin, Guang, Tao, Zerui, Zhang, Jianhai, Tanaka, Toshihisa, Zhao, Qibin

Diffusion models (DMs) based adversarial purification (AP) has shown to be the most powerful alternative to adversarial training (AT). However, these methods neglect the fact that pre-trained diffusion models themselves are not robust to adversarial

Externí odkaz: http://arxiv.org/abs/2403.16067

Zobrazit plný text záznamu

Report

Adversarial Training on Purification (AToP): Advancing Both Robustness and Generalization

Autor: Lin, Guang, Li, Chao, Zhang, Jianhai, Tanaka, Toshihisa, Zhao, Qibin

The deep neural networks are known to be vulnerable to well-designed adversarial attacks. The most successful defense technique based on adversarial training (AT) can achieve optimal robustness against particular attacks but cannot generalize well to

Externí odkaz: http://arxiv.org/abs/2401.16352

Zobrazit plný text záznamu

Report

Multi-Subdomain Adversarial Network for Cross-Subject EEG-based Emotion Recognition

Autor: Lin, Guang, Zhang, Jianhai

The individual difference between subjects is significant in EEG-based emotion recognition, resulting in the difficulty of sharing the model across subjects. Previous studies use domain adaptation algorithms to minimize the global domain discrepancy

Externí odkaz: http://arxiv.org/abs/2308.14059

Zobrazit plný text záznamu

Report

Synchronous Image-Label Diffusion Probability Model with Application to Stroke Lesion Segmentation on Non-contrast CT

Autor: Zhang, Jianhai, Wan, Tonghua, MacDonald, Ethan, Menon, Bijoy, Ganesh, Aravind, Wu, Qiu

Stroke lesion volume is a key radiologic measurement for assessing the prognosis of Acute Ischemic Stroke (AIS) patients, which is challenging to be automatically measured on Non-Contrast CT (NCCT) scans. Recent diffusion probabilistic models have sh

Externí odkaz: http://arxiv.org/abs/2307.01740

Zobrazit plný text záznamu

Report

ChatPLUG: Open-Domain Generative Dialogue System with Internet-Augmented Instruction Tuning for Digital Human

Autor: Tian, Junfeng, Chen, Hehong, Xu, Guohai, Yan, Ming, Gao, Xing, Zhang, Jianhai, Li, Chenliang, Liu, Jiayi, Xu, Wenshen, Xu, Haiyang, Qian, Qi, Wang, Wei, Ye, Qinghao, Zhang, Jiejing, Zhang, Ji, Huang, Fei, Zhou, Jingren

In this paper, we present ChatPLUG, a Chinese open-domain dialogue system for digital human applications that instruction finetunes on a wide range of dialogue tasks in a unified internet-augmented format. Different from other open-domain dialogue mo

Externí odkaz: http://arxiv.org/abs/2304.07849

Zobrazit plný text záznamu