paper-with-me

홈 › Papers

Exploiting Prefix-Tree in Structured Output Interfaces for Enhancing Jailbreak Attacking

2025-02-19 · Yanzeng Li, Yunfan Xiong, Jialun Zhong, Jinchao Zhang, Jie zhou, Lei Zou

The rise of Large Language Models (LLMs) has led to significant applications but also introduced serious security threats, particularly from jailbreak attacks that manipulate output generation. These attacks utilize prompt engineering and logit manipulation to steer models toward harmful content, prompting LLM providers to implement filtering and safety alignment strategies. We investigate LLMs' safety mechanisms and their recent applications, revealing a new threat model targeting structured output interfaces, which enable attackers to manipulate the inner logit during LLM generation, requiring only API access permissions. To demonstrate this threat model, we introduce a black-box attack framework called AttackPrefixTree (APT). APT exploits structured output interfaces to dynamically construct attack patterns. By leveraging prefixes of models' safety refusal response and latent harmful outputs, APT effectively bypasses safety measures. Experiments on benchmark datasets indicate that this approach achieves higher attack success rate than existing methods. This work highlights the urgent need for LLM providers to enhance security protocols to address vulnerabilities arising from the interaction between safety patterns and structured outputs.

📄 PDF Abstract BibTeX arXiv:2502.13527

Code (1)

lsvih/attackPrefixTree 공식 구현 pytorch

Tasks

Prompt EngineeringSafety Alignment

Similar Papers 제목 키워드 기반

Tree Training: Accelerating Agentic LLMs Training via Shared Prefix Reuse

2025-11-01 · Jinghui Wang, Shaojie Wang, Yinghan Cui, Xuxing Chen 외 arxiv

Agentic large language model (LLM) training often involves multi-turn interaction trajectories that branch into multiple execution paths due to concurrent tool use, think-mode, sub-agent, context management and other run…

Reinforcement Learning

DeFT: Decoding with Flash Tree-attention for Efficient Tree-structured LLM Inference

2024-03-30 · Jinwei Yao, Kaiqi Chen, Kexun Zhang, Jiaxuan You 외

Large language models (LLMs) are increasingly employed for complex tasks that process multiple generation calls in a tree structure with shared prefixes of tokens, including few-shot prompting, multi-step reasoning, spec…

GPU

Exploiting Tree Structure for Credit Assignment in RL Training of LLMs

2025-09-22 · Hieu Tran, Zonghai Yao, Hong Yu arxiv

Reinforcement learning improves LLM reasoning, yet sparse delayed reward over long sequences makes token-level credit assignment the key bottleneck. We study the verifiable-reward setting, where the final answer is check…

Reinforcement Learning

Evaluating Node-tree Interfaces for AI Explainability

2025-10-07 · Lifei Wang, Natalie Friedman, Chengchao Zhu, Zeshu Zhu 외 arxiv

As large language models (LLMs) become ubiquitous in workplace tools and decision-making processes, ensuring explainability and fostering user trust are critical. Although advancements in LLM engineering continue, human-…

Spend Search Where It Pays: Value-Guided Structured Sampling and Optimization for Generative Recommendation

2026-02-11 · Jie Jiang, Yangru Huang, Zeyu Wang, Changping Wang 외 arxiv

Generative recommendation via autoregressive models has unified retrieval and ranking into a single conditional generation framework. However, fine-tuning these models with Reinforcement Learning (RL) often suffers from …

Reinforcement Learning