Papers by Zhaoxia Yin
AutoBreach: Universal and Adaptive Jailbreaking with Efficient Wordplay-Guided Optimization via Multi-LLMs (2025.findings-naacl)
Copied to clipboard
| Challenge: | Existing jailbreak research exhibits limitations in universality, validity, and efficiency . Existing methods for jailbreaking LLMs have limited validity and effectiveness . |
| Approach: | They propose a black-box approach that uses wordplay-guided mapping rule sampling to create universal adversarial prompts. |
| Outcome: | The proposed method efficiently identifies security vulnerabilities across various LLMs, achieving an average success rate of over 80% with fewer than 10 queries. |
Auto-Stega: An Agent-Driven System for Lifelong Strategy Evolution in LLM-Based Text Steganography (2026.findings-acl)
Copied to clipboard
| Challenge: | prevailing methods rely on hand-crafted or pre-specified strategies and struggle to balance efficiency, imperceptibility, and security, particularly at high embedding rates. |
| Approach: | They propose an agent-driven self-evolving framework that is the first to realize self-changing steganographic strategies by automatically discovering, composing, and adapting strategies at inference time. |
| Outcome: | The proposed framework achieves 42.2% perplexity and 1.6% anti-steganalysis performance over SOTA methods at high embedding rates. |
Red Teaming Large Reasoning Models (2026.acl-long)
Copied to clipboard
| Challenge: | Large Reasoning Models (LRMs) have emerged as a powerful advancement in multi-step reasoning tasks, but they introduce safety and reliability risks, such as CoT-hijacking and prompt-induced inefficiencies. |
| Approach: | They propose a unified benchmark to assess the trustworthiness of Large Reasoning Models. |
| Outcome: | The proposed benchmark evaluates truthfulness, safety and efficiency on 26 models. |
LCO: LLM-based Constraint Optimization for Safer Agentic LLMs in Real-world Tasks (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing defense methods are insufficient to address in-context reward hacking (ICRH), where LLMs iteratively optimize their behavior to maximize proxy objectives, resulting in harmful side effects. |
| Approach: | They propose a framework that reduces in-context reward hacking (ICRH) through repeated interactions with the environment. |
| Outcome: | The proposed framework reduces ICRH without model fine-tuning while maintaining task performance. |