Minghui Li

Venues

Frequent coauthors

Papers

1. Dual-View Inference Attack: Machine Unlearning Amplifies Privacy Exposure
   AAAI 2026 ·

   Lulu Xue

   DBLP profile ↗ORCID search ↗
2. AdvEDM: Fine-grained Adversarial Attack against VLM-based Embodied Agents
   NeurIPS 2025 ·

   Yichen Wang

   DBLP profile ↗ORCID search ↗
3. BadRobot: Jailbreaking Embodied LLM Agents in the Physical World
   ICLR 2025 ·

   Hangtao Zhang

   DBLP profile ↗ORCID search ↗
4. Breaking Barriers in Physical-World Adversarial Examples: Improving Robustness and Transferability via Robust Feature
   AAAI 2025 ·

   Yichen Wang

   DBLP profile ↗ORCID search ↗
5. Detecting and Corrupting Convolution-based Unlearnable Examples
   AAAI 2025 · Minghui Li
6. Improving Generalization of Universal Adversarial Perturbation via Dynamic Maximin Optimization
   AAAI 2025 ·

   Yechao Zhang

   DBLP profile ↗ORCID search ↗
7. Test-Time Backdoor Detection for Object Detection Models
   CVPR 2025 ·

   Hangtao Zhang

   DBLP profile ↗ORCID search ↗
8. Transferable Direct Prompt Injection via Activation-Guided MCMC Sampling
   EMNLP 2025 · Minghui Li
9. Uncertainty-Aware Metabolic Stability Prediction with Dual-View Contrastive Learning
   ECML-PKDD 2025 ·

   Peijin Guo

   DBLP profile ↗ORCID search ↗
10. DarkFed: A Data-Free Backdoor Attack in Federated Learning
    IJCAI 2024 · Minghui Li
11. DarkSAM: Fooling Segment Anything Model to Segment Nothing
    NeurIPS 2024 ·

    Ziqi Zhou

    DBLP profile ↗ORCID search ↗
12. Transferable Adversarial Facial Images for Privacy Protection
    ACM MM 2024 · Minghui Li
13. Unlearnable 3D Point Clouds: Class-wise Transformation Is All You Need
    NeurIPS 2024 ·

    Xianlong Wang

    DBLP profile ↗ORCID search ↗