Papers by Zhun Wang
AGENTVIGIL: Automatic Black-Box Red-teaming for Indirect Prompt Injection against LLM Agents (2025.findings-emnlp)
Copied to clipboard
Zhun Wang, Vincent Siu, Zhe Ye, Tianneng Shi, Yuzhou Nie, Xuandong Zhao, Chenguang Wang, Wenbo Guo, Dawn Song
| Challenge: | AGENTVIGIL is a black-box optimization framework to exploit indirect prompt injection vulnerabilities . indirect prompts compromise the core of LLM agents by manipulating contextual information rather than direct user prompts. |
| Approach: | They propose a black-box optimization framework to exploit indirect prompt injection vulnerabilities . they use a Monte Carlo tree-based algorithm to iteratively refine inputs . |
| Outcome: | The proposed framework achieves 71% and 70% success rates against two public benchmarks . |
FakeSV-VLM: Taming VLM for Detecting Fake Short-Video News via Progressive Mixture-Of-Experts Adapter (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for detecting fake news videos fall short due to lack of knowledge to verify the news is real or not. |
| Approach: | They propose a VLM-based framework for detecting fake news on short video platforms . they design four experts tailored to handle each scenario and integrate them into VLM . |
| Outcome: | The proposed framework outperforms current state-of-the-art models on two benchmark datasets. |
COSMIC: Generalized Refusal Direction Identification in LLM Activations (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods depend on predefined refusal templates detectable in output tokens or manual review. |
| Approach: | They propose a framework that optimally identifies steering directions and target layers using cosine similarity, entirely independent of output text. |
| Outcome: | The proposed framework achieves comparable steering effectiveness without any prior knowledge or assumptions of a model’s refusal behavior such as the use of certain refusal tokens. |
Synergistic Weak-Strong Collaboration by Aligning Preferences (2025.acl-long)
Copied to clipboard
Yizhu Jiao, Xuchao Zhang, Zhaoyang Wang, Yubo Ma, Zhun Deng, Rujia Wang, Chetan Bansal, Saravan Rajmohan, Jiawei Han, Huaxiu Yao
| Challenge: | Current Large Language Models excel in general reasoning yet struggle with specialized tasks requiring proprietary or domain-specific knowledge. |
| Approach: | They propose a collaborative framework that pairs a specialized weak model with a general strong model to optimize collaboration. |
| Outcome: | The proposed framework outperforms each model alone by leveraging complementary strengths. |