本日紹介した論文の一覧 Against All Odds: Overcoming Typology, Script, and Language Confusion in Multilingual Embedding Inversion Attacks http://arxiv.org/abs/2408.11749v1 Efficient Detection of Toxic Prompts in Large Language Models http://arxiv.org/abs/2408.11727v1 CIPHER: Cybersecurity Intelligent Penetration-testing Helper for Ethical Researcher http://arxiv.org/abs/2408.11650v1 Private Counting of Distinct Elements in the Turnstile Model and Extensions http://arxiv.org/abs/2408.11637v1 Confidential Computing on Heterogeneous Systems: Survey and Implications http://arxiv.org/abs/2408.11601v1 Large Language Models are Good Attackers: Efficient and Stealthy Textual Backdoor Attacks http://arxiv.org/abs/2408.11587v1 Characterizing the Evolution of Psychological Tactics and Techniques Exploited by Malicious Emails http://arxiv.org/abs/2408.11586v1 Characterizing the Evolution of Psychological Factors Exploited by Malicious Emails http://arxiv.org/abs/2408.11584v1 Constructions of Efficiently Implementable Boolean functions Possessing High Nonlinearity and Good Resistance to Algebraic Attacks http://arxiv.org/abs/2408.11583v1 Security Evaluation in Software-Defined Networks http://arxiv.org/abs/2408.11486v1 A Practical Trigger-Free Backdoor Attack on Neural Networks http://arxiv.org/abs/2408.11444v1 EEG-Defender: Defending against Jailbreak through Early Exit Generation of Large Language Models http://arxiv.org/abs/2408.11308v1 Correlation Analysis of Adversarial Attack in Time Series Classification http://arxiv.org/abs/2408.11264v1 Privacy-Preserving Data Management using Blockchains http://arxiv.org/abs/2408.11263v1 なお、ポッドキャスト内で紹介する内容は、各論文の概要を日本語で解説したもので、論文概要の著作権は論文著者に帰属します。]]

Ask this episode

“What did the council decide on item 42?” — answered with the speaker and the moment.

Transcript

Related