Lune

Profiles2021-2026

Yang Zhang

CISPA Helmholtz Center for Information Security

20Papers
15Last author
39Unique co-authors
0First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers20

  1. BadTV: Unveiling Backdoor Threats in Third-Party Task VectorsAuthor 5 of 10CCS 2026 · 2 citations
  2. Rethinking Assessments of Prompt Injection AttacksAuthor 4 of 4ACL Findings 2026 · 1 citation
  3. DE-CLIP: Few-Shot Anomaly Detection via Difference-Guided Embedding EditingAuthor 4 of 4ACL 2026
  4. InferPilot: Autonomous Inference Attacks Against ML Services With LLM-Based AgentsAuthor 5 of 5ACL Findings 2026
  5. Peering Behind the Shield: Guardrail Identification in Large Language ModelsAuthor 5 of 5ACL Findings 2026
  6. Reward Yourself: Efficient Self Rewards for Trustworthy SamplingAuthor 4 of 4ACL Findings 2026
  7. GPTracker: A Large-Scale Measurement of Misused GPTsAuthor 4 of 4S&P 2025
  8. On the Effectiveness of Prompt Stealing Attacks on In-the-Wild PromptsAuthor 5 of 5S&P 2025
  9. Towards Understanding Unsafe Video GenerationAuthor 3 of 4NDSS 2025
  10. Understanding Data Importance in Machine Learning Attacks: Does Valuable Data Pose Greater Harm?Author 3 of 3NDSS 2025
  11. ZeroFake: Zero-Shot Detection of Fake Images Generated and Edited by Text-to-Image Generation ModelsAuthor 4 of 4CCS 2024 · 8 citations
  12. The Death and Life of Great Prompts: Analyzing the Evolution of LLM Prompts from the Structural PerspectiveAuthor 6 of 6EMNLP 2024 · 6 citations
  13. BadMerging: Backdoor Attacks Against Model MergingAuthor 5 of 6CCS 2024 · 5 citations
  14. Reconstruct Your Previous Conversations! Comprehensively Investigating Privacy Leakage Risks in Conversations with GPT ModelsAuthor 3 of 3EMNLP 2024 · 3 citations
  15. ModSCAN: Measuring Stereotypical Bias in Large Vision-Language Models from Vision and Language ModalitiesAuthor 6 of 6EMNLP 2024 · 1 citation
  16. Pseudo Label-Guided Model Inversion Attack via Conditional Generative Adversarial NetworkAuthor 6 of 6AAAI 2023 · 57 citations
  17. NOTABLE: Transferable Backdoor Attacks Against Prompt-based NLP ModelsAuthor 4 of 5ACL 2023 · 17 citations
  18. Can't Steal? Cont-Steal! Contrastive Stealing Attacks Against Image EncodersAuthor 5 of 5CVPR 2023
  19. Is Adversarial Training Really a Silver Bullet for Mitigating Data Poisoning?Author 6 of 6ICLR 2023
  20. "Go eat a bat, Chang!": On the Emergence of Sinophobic Behavior on Web Communities in the Face of COVID-19Author 6 of 7WWW 2021 · 92 citations
Dusk over the sea between two cliffs drawn in fine vertical lines