基於生成式模型之語音及文字惡意攻擊的防禦技術;Defense Techniques against Adversarial Attacks on Speech and Text Based on Generative Models

NCUIR > College of Electrical Engineering & Computer Science > Department of Computer Science and information Engineering > Research Project > Item 987654321/96899

Please use this identifier to cite or link to this item: https://ir.lib.ncu.edu.tw/handle/987654321/96899

Title:	基於生成式模型之語音及文字惡意攻擊的防禦技術;Defense Techniques against Adversarial Attacks on Speech and Text Based on Generative Models
Authors:	王家慶;呂俊賢
Contributors:	國立中央大學資訊工程學系
Keywords:	惡意攻擊;惡意攻擊的防禦;後門攻擊;擴散模型;生成式模型;Adversarial Attack;Adversarial Defense;Backdoor Attack;Diffusion Model;Generative Model
Date:	2025-07-31
Issue Date:	2025-08-07 17:18:23 (UTC+8)
Publisher:	國家科學及技術委員會(本會)
Abstract:	本計畫將建立語音及文字系統的防禦技術框架，並探索大語言模型在系統安全中的創新應用，為深度學習模型的實用性與安全性提供全面支撐，促進該領域的長遠發展。
Relation:	財團法人國家實驗研究院科技政策研究與資訊中心
Appears in Collections:	[Department of Computer Science and information Engineering] Research Project

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	4	View/Open

社群 sharing