Yige Li

Venues

Frequent coauthors

Papers

1. BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents
   ACL 2026 ·

   Yunhao Feng

   DBLP profile ↗ORCID search ↗
2. Unleashing the Unseen: Harnessing Benign Datasets for Jailbreaking Large Language Models
   EACL 2026 ·

   Wei Zhao

   DBLP profile ↗ORCID search ↗
3. Anyattack: Towards Large-scale Self-supervised Adversarial Attacks on Vision-language Models
   CVPR 2025 ·

   Jiaming Zhang

   DBLP profile ↗ORCID search ↗
4. Backdoor Token Unlearning: Exposing and Defending Backdoors in Pretrained Language Models
   AAAI 2025 ·

   Peihai Jiang

   DBLP profile ↗ORCID search ↗
5. BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models
   NeurIPS 2025 · Yige Li
6. BlueSuffix: Reinforced Blue Teaming for Vision-Language Models Against Jailbreak Attacks
   ICLR 2025 ·

   Yunhan Zhao

   DBLP profile ↗ORCID search ↗
7. CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularization
   ICML 2025 ·

   Nay Myat Min

   DBLP profile ↗ORCID search ↗
8. Detecting Backdoor Samples in Contrastive Language Image Pretraining
   ICLR 2025 ·

   Hanxun Huang

   DBLP profile ↗ORCID search ↗
9. Do Influence Functions Work on Large Language Models?
   EMNLP 2025 ·

   Zhe Li

   DBLP profile ↗ORCID search ↗
10. Memory Injection Attacks on LLM Agents via Query-Only Interaction
    NeurIPS 2025 ·

    Shen Dong

    DBLP profile ↗ORCID search ↗
11. X-Transfer Attacks: Towards Super Transferable Adversarial Attacks on CLIP
    ICML 2025 ·

    Hanxun Huang

    DBLP profile ↗ORCID search ↗
12. Zero-Shot Defense Against Toxic Images via Inherent Multimodal Alignment in LVLMs
    EMNLP 2025 ·

    Wei Zhao

    DBLP profile ↗ORCID search ↗
13. Defending Large Language Models Against Jailbreak Attacks via Layer-specific Editing
    EMNLP 2024 ·

    Wei Zhao

    DBLP profile ↗ORCID search ↗