Description

本日紹介した論文の一覧

• SPATA: Systematic Pattern Analysis for Detailed and Transparent Data
  Cards
  http://arxiv.org/abs/2509.26640v1

• Are Robust LLM Fingerprints Adversarially Robust?
  http://arxiv.org/abs/2509.26598v1

• DeepProv: Behavioral Characterization and Repair of Neural Networks via
  Inference Provenance Graph Analysis
  http://arxiv.org/abs/2509.26562v1

• Explainable and Resilient ML-Based Physical-Layer Attack Detectors
  http://arxiv.org/abs/2509.26530v1

• Logic Solver Guided Directed Fuzzing for Hardware Designs
  http://arxiv.org/abs/2509.26509v1

• SeedPrints: Fingerprints Can Even Tell Which Seed Your Large Language
  Model Was Trained From
  http://arxiv.org/abs/2509.26404v1

• Exact Bias of Linear TRNG Correctors -- Spectral Approach
  http://arxiv.org/abs/2509.26393v1

• SoK: Systematic analysis of adversarial threats against deep learning
  approaches for autonomous anomaly detection systems in SDN-IoT networks
  http://arxiv.org/abs/2509.26350v1

• Strong random unitaries and fast scrambling
  http://arxiv.org/abs/2509.26310v1

• Stealthy Yet Effective: Distribution-Preserving Backdoor Attacks on
  Graph Classification
  http://arxiv.org/abs/2509.26032v1

• The Impact of Scaling Training Data on Adversarial Robustness
  http://arxiv.org/abs/2509.25927v1

• Better Privilege Separation for Agents by Restricting Data Types
  http://arxiv.org/abs/2509.25926v1

• STAC: When Innocent Tools Form Dangerous Chains to Jailbreak LLM Agents
  http://arxiv.org/abs/2509.25624v1

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。