2026
[EMNLP] INTENT-AS-A-TOOL Makes it Easy to Track Agentic Misalignment
[EMNLP] The Persona-Switch Backdoor: From Payload Retrieval to Trigger-Gated Persona-State Selection
[EMNLP] Auditing Chinese Web-scale Corpora via Sampled BPE Token Statistics
[ACM MM] Beyond Retrieval: Improving Evidence Quality for LLM-based Multimodal Fact-Checking
[ECCV] GlassGS: Geometry and Concept-Aware 3D Gaussian Splatting for Reflective Enclosures
[TIFS] Towards Trustworthy Dynamic Facial Expression Recognition via Information Bottleneck Modeling
[TPAMI] Exploring Security Vulnerabilities in Multilingual Speech Translation Systems via Deceptive Inputs
[Project Page]
[Project Page]
[USENIX Security] When Memory Becomes a Vulnerability: Towards Multi-turn Jailbreak Attacks against Text-to-Image Generation Systems
[CCS] Token Buncher: Shielding LLMs from Harmful Reinforcement Learning Fine-Tuning
[USENIX Security] One Bad Token Spoils the Barrel: Assessment, Detection, and Remediation of Glitch Tokens in Large Language Models
[ICML] When Search Goes Wrong: Red-Teaming Web-Augmented Large Language Models
[ICML] OBJVanish: Prompt-Driven Generation of Physically Realisable 3D LiDAR-Invisible Objects
[ICML] State-Dependent Safety Failures in Multi-Turn Language Model Interaction
[ACL Findings] CORBA: Contagious Recursive Blocking Attacks on Multi-Agent Systems Based on Large Language Models
[ICRA] SURE: Semi-dense Uncertainty-REfined Feature Matching
2025
[Information Science] Versatile and harmless deepfake proactive forensics via conditional watermarking
[TIFS] SSD: A State-based Stealthy Backdoor Attack For IMU/GNSS Navigation System in UAV Route Planning
[Agent4Science] Visible Yet Unreadable: A Systematic Blind Spot of Vision-Language Models Across Writing Systems
Spotlight [Code]
Spotlight [Code]
[NeurIPS] DepthVanish: Optimizing Adversarial Interval Structures for Stereo-Depth-Invisible Patches
[TPAMI] FaceTracer: Unveiling Source Identities from Swapped Face Images and Videos for Fraud Prevention
[USENIX Security] DiffLoc: WiFi Hidden Camera Localization Based on Electromagnetic Diffraction
[Code]
[Code]
[CCS] SafeGuider: Robust and Practical Content Safety Control for Text-to-Image Models
[Project Page]
[Project Page]
[ICME] A Watermark Updating Framework for Multi-stage Image Content Distribution
One of the Top 15 Papers
One of the Top 15 Papers
[S&P] CamLoPA: A Hidden Wireless Camera Localization Framework via Signal Propagation Path Analysis
[Code] [Video]
[Code] [Video]
[S&P] Towards Reliable Verification of Unauthorized Data Usage in Personalized Text-to-Image Diffusion Models
[Code]
[Code]
[USENIX Security] When Translators Refuse to Translate: A Novel Attack to Speech Translation Systems
[USENIX Security] The Ghost Navigator: Revisiting the Hidden Vulnerability of Localization in Autonomous Driving
[TIFS] Turning Your Strength into Watermark: Watermarking Large Language Model via Knowledge Injection
[AAAI] An LLM-empowered Adaptive Evolutionary Algorithm For Multi-Component Deep Learning Systems
Oral
Oral
2024
[Information Fusion] Security Analysis and Adaptive False Data Injection against MultiSensor Fusion Localization for Autonomous Driving
[MM] FacialPulse: An Efficient RNN-based Depression Detection via Temporal Facial Landmarks
Oral (3.97%) [Code]
Oral (3.97%) [Code]
[ICML] AquaLoRA: Toward White-box Protection for Customized Stable Diffusion Models via Watermark LoRA
[Code]
[Code]
[CCS] GenderCARE: A Comprehensive Framework for Assessing and Reducing Gender Bias in Large Language Models
Distinguished Artifact Award [Project Page]
Distinguished Artifact Award [Project Page]
[AAAI] Personalization as a Shortcut for Few-Shot Backdoor Attack against Text-to-Image Diffusion Models
[Code]
[Code]
[ICASSP] Attribute-Aware Head Swapping Guided by 3d Modeling
2023
[Springer Book] Digital Watermarking for Machine Learning Models - Chapter 6: Protecting Image Processing Networks via Model Watermarking
[AAAI] Pseudo Label-Guided Model Inversion Attack via Conditional Generative Adversarial Network
[Code]
[Code]