文献データベース

The Early Bird Catches the Leak: Unveiling Timing Side Channels in LLM Serving Systems

Authors: Linke Song, Zixuan Pang, Wenhao Wang, Zihao Wang, XiaoFeng Wang, Hongbo Chen, Wei Song, Yier Jin, Dan Meng, Rui Hou | Published: 2024-09-30 | Updated: 2025-08-13
セキュリティ保証
プロンプトリーキング
攻撃戦略分析

Infighting in the Dark: Multi-Label Backdoor Attack in Federated Learning

Authors: Ye Li, Yanchao Zhao, Chengcheng Zhu, Jiale Zhang | Published: 2024-09-29 | Updated: 2025-03-22
IDマッピングの構築
バックドアモデルの検知
敵対的攻撃

OnePath: Efficient and Privacy-Preserving Decision Tree Inference in the Cloud

Authors: Shuai Yuan, Hongwei Li, Xinyuan Qian, Guowen Xu | Published: 2024-09-28 | Updated: 2025-07-21
プライバシーと最適化
性能評価手法
暗号化手法

Enhancing Robustness of Graph Neural Networks through p-Laplacian

Authors: Anuj Kumar Sirohi, Subhanu Halder, Kabir Kumar, Sandeep Kumar | Published: 2024-09-27
最適化問題
防御手法

System-Level Defense against Indirect Prompt Injection Attacks: An Information Flow Control Perspective

Authors: Fangzhou Wu, Ethan Cecchetti, Chaowei Xiao | Published: 2024-09-27 | Updated: 2024-10-10
LLMセキュリティ
プロンプトインジェクション
実行トレースの妨害

Code Vulnerability Repair with Large Language Model using Context-Aware Prompt Tuning

Authors: Arshiya Khan, Guannan Liu, Xing Gao | Published: 2024-09-27 | Updated: 2025-06-11
コード脆弱性修復
セキュリティコンテキスト統合
大規模言語モデル

An Adversarial Perspective on Machine Unlearning for AI Safety

Authors: Jakub Łucki, Boyi Wei, Yangsibo Huang, Peter Henderson, Florian Tramèr, Javier Rando | Published: 2024-09-26 | Updated: 2025-04-10
プロンプトインジェクション
安全性アライメント
機械学習の忘却

Weak-to-Strong Backdoor Attack for Large Language Models

Authors: Shuai Zhao, Leilei Gan, Zhongliang Guo, Xiaobao Wu, Luwei Xiao, Xiaoyu Xu, Cong-Duy Nguyen, Luu Anh Tuan | Published: 2024-09-26 | Updated: 2024-10-13
バックドア攻撃
プロンプトインジェクション

MoJE: Mixture of Jailbreak Experts, Naive Tabular Classifiers as Guard for Prompt Attacks

Authors: Giandomenico Cornacchia, Giulio Zizzo, Kieran Fraser, Muhammad Zaid Hameed, Ambrish Rawat, Mark Purcell | Published: 2024-09-26 | Updated: 2024-10-04
ガードレール手法
コンテンツモデレーション
プロンプトインジェクション

A novel application of Shapley values for large multidimensional time-series data: Applying explainable AI to a DNA profile classification neural network

Authors: Lauren Elborough, Duncan Taylor, Melissa Humphries | Published: 2024-09-26
アルゴリズム
ウォーターマーキング
評価手法