Description

本日紹介した論文の一覧

• Bag of Tricks: Benchmarking of Jailbreak Attacks on LLMs
  http://arxiv.org/abs/2406.09324v1

• JailbreakEval: An Integrated Toolkit for Evaluating Jailbreak Attempts
  Against Large Language Models
  http://arxiv.org/abs/2406.09321v1

• EncCluster: Scalable Functional Encryption in Federated Learning through
  Weight Clustering and Probabilistic Filters
  http://arxiv.org/abs/2406.09152v1

• Privacy Aware Memory Forensics
  http://arxiv.org/abs/2406.09005v1

• A Passwordless MFA Utlizing Biometrics, Proximity and Contactless
  Communication
  http://arxiv.org/abs/2406.09000v1

• Weaponizing Disinformation Against Critical Infrastructures
  http://arxiv.org/abs/2406.08963v1

• Beyond the Calibration Point: Mechanism Comparison in Differential
  Privacy
  http://arxiv.org/abs/2406.08918v1

• Improving Adversarial Robustness via Feature Pattern Consistency
  Constraint
  http://arxiv.org/abs/2406.08829v1

• Interpretable Temporal Class Activation Representation for Audio
  Spoofing Detection
  http://arxiv.org/abs/2406.08825v1

• StructuralSleight: Automated Jailbreak Attacks on Large Language Models
  Utilizing Uncommon Text-Encoded Structure
  http://arxiv.org/abs/2406.08754v1

• RL-JACK: Reinforcement Learning-powered Black-box Jailbreaking Attack
  against LLMs
  http://arxiv.org/abs/2406.08725v1

• TikTag: Breaking ARM's Memory Tagging Extension with Speculative
  Execution
  http://arxiv.org/abs/2406.08719v1

• When LLM Meets DRL: Advancing Jailbreaking Efficiency via DRL-guided
  Search
  http://arxiv.org/abs/2406.08705v1

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。