Description

本日紹介した論文の一覧

• Private Geometric Median in Nearly-Linear Time
  http://arxiv.org/abs/2505.20189v1

• Eradicating the Unseen: Detecting, Exploiting, and Remediating a Path
  Traversal Vulnerability across GitHub
  http://arxiv.org/abs/2505.20186v1

• Exposing Go's Hidden Bugs: A Novel Concolic Framework
  http://arxiv.org/abs/2505.20183v1

• Engineering Trustworthy Machine-Learning Operations with Zero-Knowledge
  Proofs
  http://arxiv.org/abs/2505.20136v1

• TrojanStego: Your Language Model Can Secretly Be A Steganographic
  Privacy Leaking Agent
  http://arxiv.org/abs/2505.20118v1

• Transformer in Protein: A Survey
  http://arxiv.org/abs/2505.20098v1

• DFIR-Metric: A Benchmark Dataset for Evaluating Large Language Models in
  Digital Forensics and Incident Response
  http://arxiv.org/abs/2505.19973v1

• Differential Privacy Analysis of Decentralized Gossip Averaging under
  Varying Threat Models
  http://arxiv.org/abs/2505.19969v1

• Novel Loss-Enhanced Universal Adversarial Patches for Sustainable
  Speaker Privacy
  http://arxiv.org/abs/2505.19951v1

• Evaluating AI cyber capabilities with crowdsourced elicitation
  http://arxiv.org/abs/2505.19915v1

• Deconstructing Obfuscation: A four-dimensional framework for evaluating
  Large Language Models assembly code deobfuscation capabilities
  http://arxiv.org/abs/2505.19887v1

• CPA-RAG:Covert Poisoning Attacks on Retrieval-Augmented Generation in
  Large Language Models
  http://arxiv.org/abs/2505.19864v1

• One Surrogate to Fool Them All: Universal, Transferable, and Targeted
  Adversarial Attacks with CLIP
  http://arxiv.org/abs/2505.19840v1

• Poison in the Well: Feature Embedding Disruption in Backdoor Attacks
  http://arxiv.org/abs/2505.19821v1

• What Really Matters in Many-Shot Attacks? An Empirical Study of
  Long-Context Vulnerabilities in LLMs
  http://arxiv.org/abs/2505.19773v1

• A Comprehensive Real-World Assessment of Audio Watermarking Algorithms:
  Will They Survive Neural Codecs?
  http://arxiv.org/abs/2505.19663v1

• STOPA: A Database of Systematic VariaTion Of DeePfake Audio for Open-Set
  Source Tracing and Attribution
  http://arxiv.org/abs/2505.19644v1

• Weak-Jamming Detection in IEEE 802.11 Networks: Techniques, Scenarios
  and Mobility
  http://arxiv.org/abs/2505.19633v1

• Language of Network: A Generative Pre-trained Model for Encrypted
  Traffic Comprehension
  http://arxiv.org/abs/2505.19482v1

• An Empirical Study of JavaScript Inclusion Security Issues in Chrome
  Extensions
  http://arxiv.org/abs/2505.19456v1

• Structure Disruption: Subverting Malicious Diffusion-Based Inpainting
  via Self-Attention Query Perturbation
  http://arxiv.org/abs/2505.19425v1

• CoTGuard: Using Chain-of-Thought Triggering for Copyright Protection in
  Multi-Agent LLM Systems
  http://arxiv.org/abs/2505.19405v1

• VADER: A Human-Evaluated Benchmark for Vulnerability Assessment,
  Detection, Explanation, and Remediation
  http://arxiv.org/abs/2505.19395v1

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。