本日紹介した論文の一覧 What If, But Privately: Private Counterfactual Retrieval http://arxiv.org/abs/2508.03681v1 MalFlows: Context-aware Fusion of Heterogeneous Flow Semantics for Android Malware Detection http://arxiv.org/abs/2508.03588v1 Intrusion Detection in Heterogeneous Networks with Domain-Adaptive Multi-Modal Learning http://arxiv.org/abs/2508.03517v1 Unravelling the Probabilistic Forest: Arbitrage in Prediction Markets http://arxiv.org/abs/2508.03474v1 Smart Car Privacy: Survey of Attacks and Privacy Issues http://arxiv.org/abs/2508.03413v1 When Good Sounds Go Adversarial: Jailbreaking Audio-Language Models with Benign Inputs http://arxiv.org/abs/2508.03365v1 From Legacy to Standard: LLM-Assisted Transformation of Cybersecurity Playbooks into CACAO Format http://arxiv.org/abs/2508.03342v1 Bidirectional TLS Handshake Caching for Constrained Industrial IoT Scenarios http://arxiv.org/abs/2508.03321v1 BDFirewall: Towards Effective and Expeditiously Black-Box Backdoor Defense in MLaaS http://arxiv.org/abs/2508.03307v1 BadBlocks: Low-Cost and Stealthy Backdoor Attacks Tailored for Text-to-Image Diffusion Models http://arxiv.org/abs/2508.03221v1 WiFinger: Fingerprinting Noisy IoT Event Traffic Using Packet-level Sequence Matching http://arxiv.org/abs/2508.03151v1 Protecting Small Organizations from AI Bots with Logrip: Hierarchical IP Hashing http://arxiv.org/abs/2508.03130v1 Attack the Messages, Not the Agents: A Multi-round Adaptive Stealthy Tampering Framework for LLM-MAS http://arxiv.org/abs/2508.03125v1 VFLAIR-LLM: A Comprehensive Framework and Benchmark for Split Learning of LLMs http://arxiv.org/abs/2508.03097v1 T2UE: Generating Unlearnable Examples from Text Descriptions http://arxiv.org/abs/2508.03091v1 Untraceable DeepFakes via Traceable Fingerprint Elimination http://arxiv.org/abs/2508.03067v1 Lightweight Fault Detection Architecture for NTT on FPGA http://arxiv.org/abs/2508.03062v1 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