Wireless Networks Security and Privacy

• Course Information

• Semester: 2026 Autumn
• Time: 13:25-17:00 (6.7.8.9) Thursday
• Location: Yuquan Campus, #4 Teaching Building, Rm204
• TA: Yu Wang, wy_chaser@zju.edu.cn

• Schedule (Update: 2026.10.8)

Week	Course/Presentation Topic	Presenter
Week 1 17, Sep	Lecture 1. Course introduction & logistics Lecture 2. Wireless security basics & threat models	Lecturer
Week 2 24, Sep	Leave	Lecturer
Week 3 8, Oct	Lecture 3 - Information security & privacy Lecture 4 - AI security	Lecturer
Week 5 15, Oct	Lecture 5 - PHY layer Lecture 6 - Link layer	Lecturer
Week 6 22, Oct	Lecture 7 - Network layer Lecture 8 - Transport Layer	Lecturer
Week 7 29, Oct	Lecture 9 - Cross-Layer Attack & Defense Lecture 10 - Topic 1 - IoT security	Lecturer
Week 8 5, Nov	Paper Presentation	Student
Make-up class 5, Nov Periods 11-14	Paper Presentation	Student

• Group and Presentation Guidelines

• Group: Form a team individually and fill in the shared document in the group to complete the selection of the paper.
• Student presentation:
  1. 1 students as a group
  2. chooses paper from the list I provide
  3. Presentations last for 10 mins(paper) + Q&A (3 mins)
  4. Make slides yourself
• Paper source:
  1. Big 4: USENIXS Security/ NDSS/ ACM CCS/ IEEE Oakland (S&P)
  2. CCF-A: top-tier AI conferences

• Candidate papers from me (Update: 2026.10.8)
Session 1: Embodied AI

1. SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control（Science Robotics 2026）
2. EgoScale: Scaling Dexterous Manipulation with Diverse Egocentric Human Data（Technical Report / arXiv 2026）
3. Emergence of Human to Robot Transfer in Vision-Language-Action Models（RSS 2026）
4. Steerable Vision-Language-Action Policies for Embodied Reasoning and Hierarchical Control（RSS 2026）
5. On Robustness of Vision-Language-Action Model against Multi-Modal Perturbations（ICLR 2026）
6. RAG-Diff: Adapting Diffusion Policies to Dynamic Constraints with Retrieval-Augmented Guidance（RSS 2026）
7. RICL: Adding In-Context Adaptability to Pre-Trained Vision-Language-Action Models（CoRL 2025）
8. RoboMME: Benchmarking and Understanding Memory for Robotic Generalist Policies（ICML 2026 — Oral）
9. AT-VLA: Adaptive Tactile Injection for Enhanced Feedback Reaction in Vision-Language-Action Models（CVPR 2026 — Oral）
10. Latent Reasoning VLA (LaRA-VLA): Latent Thinking and Prediction for Vision-Language-Action Models（ICML 2026）

Session 2: World Models

11. Self-Improving Robot Policy with Compositional World Model（RSS 2026）
12. Collaborating Visual and Parameter Spaces for Consistent Long-Horizon Embodied World Model（RSS 2026）
13. Interactive World Simulator for Robot Policy Training and Evaluation（RSS 2026）
14. Generalizing Safety Beyond Collision-Avoidance via Latent-Space Reachability Analysis（RSS 2025）
15. Uncertainty-aware Latent Safety Filters for Avoiding Out-of-Distribution Failures (UNISafe)（CoRL 2025）
16. Visuo-Tactile World Models（ICLR 2026）
17. SafeDrive: Fine-Grained Safety Reasoning for End-to-End Driving in a Sparse World（CVPR 2026 — Highlight）
18. VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model（ECCV 2026）
19. Chain of World: World Model Thinking in Latent Motion（CVPR 2026）
20. V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning（Technical Report / arXiv 2025）

Session 3: Physical AI

21. HAMLOCK: HArdware-Model LOgically Combined attacK（USENIX Security 2026）
22. The Person Behind the Sound: Demystifying Audio Private Attribute Profiling via Multimodal Large Language Models（IEEE S&P 2026）
23. Rethinking Camera Choice: An Empirical Study on Fisheye Camera Properties in Robotic Manipulation（CVPR 2026）
24. VoiceRadar: Voice Deepfake Detection using Micro-Frequency and Compositional Analysis（NDSS 2026）
25. Shadows Don’t Lie and Lines Can’t Bend! Generative Models Don’t Know Projective Geometry...for Now（CVPR 2024）
26. Point Cloud as a Foreign Language for Multi-modal Large Language Model（CVPR 2026）
27. Unlocking the Value of Text: Event-Driven Reasoning and Multi-Level Alignment for Time Series Forecasting（ICLR 2026）
28. MERLIN: Building Low-SNR Robust Multimodal LLMs for Electromagnetic Signals（CVPR 2026）
29. CamPI: Physical Adversarial Examples through Camera Power Signal Injection（CVPR 2026）

Session 4: Foundation Models

30. SoK: Attack and Defense Landscape of Agentic AI Systems（USENIX Security 2026）
31. Q-MLLM: Vector Quantization for Robust Multimodal Large Language Model Security（NDSS 2026）
32. Refusal Is Not an Option: Unlearning Safety Alignment of Large Language Models（USENIX Security 2025）
33. FirmAgent: Leveraging Fuzzing to Assist LLM Agents with IoT Firmware Vulnerability Discovery（NDSS 2026）
34. Reference Attack: A New Cross-Modal Jailbreaking Attack against Multimodal Large Language Models（ACL 2026）
35. Internal Planning in Language Models: Characterizing Horizon and Branch Awareness（ICLR 2026）
36. REAL: Reading Out Transformer Activations for Precise Localization in Language Model Steering（ICLR 2026）
37. Scaling Laws Meet Model Architecture: Toward Inference-Efficient LLMs（ICLR 2026）
38. Training Large Language Models to Reason in a Continuous Latent Space（COLM 2025）
39. VL-JEPA: Joint Embedding Predictive Architecture for Vision-Language（ICLR 2026）

• Schedules and Materials (Update: 2026.10.8)

• Presentation and Project Schedule: Please find your presentation order in the attchment: Scoring Sheet
• Course slides: Lecture 1
• Course slides: Lecture 2