Description

本日紹介した論文の一覧

• Uncertain Location Transmitter and UAV-Aided Warden Based LEO Satellite
  Covert Communication Systems
  http://arxiv.org/abs/2504.10347v1

• Shield Bash: Abusing Defensive Coherence State Retrieval to Break Timing
  Obfuscation
  http://arxiv.org/abs/2504.10318v1

• RealHarm: A Collection of Real-World Language Model Application Failures
  http://arxiv.org/abs/2504.10277v1

• Universally Composable Commitments with Communicating Malicious
  Physically Uncloneable Functions
  http://arxiv.org/abs/2504.10120v1

• Benchmarking Practices in LLM-driven Offensive Security: Testbeds,
  Metrics, and Experiment Design
  http://arxiv.org/abs/2504.10112v1

• Quantifying Privacy Leakage in Split Inference via Fisher-Approximated
  Shannon Information Analysis
  http://arxiv.org/abs/2504.10016v1

• Do We Really Need Curated Malicious Data for Safety Alignment in
  Multi-modal Large Language Models?
  http://arxiv.org/abs/2504.10000v1

• EthCluster: An Unsupervised Static Analysis Method for Ethereum Smart
  Contract
  http://arxiv.org/abs/2504.09977v1

• Proofs of Useful Work from Arbitrary Matrix Multiplication
  http://arxiv.org/abs/2504.09971v1

• Revisiting the attacker's knowledge in inference attacks against
  Searchable Symmetric Encryption
  http://arxiv.org/abs/2504.09879v1

• StruPhantom: Evolutionary Injection Attacks on Black-Box Tabular Agents
  Powered by Large Language Models
  http://arxiv.org/abs/2504.09841v1

• SafeSpeech: Robust and Universal Voice Protection Against Malicious
  Speech Synthesis
  http://arxiv.org/abs/2504.09839v1

• An Investigation of Large Language Models and Their Vulnerabilities in
  Spam Detection
  http://arxiv.org/abs/2504.09776v1

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。