Zizhuang Deng

Guangyu Gong, Zizhuang Deng

Large Language Model (LLM) agents are increasingly integrated into critical systems, leveraging external tools to interact with the real world. However, this capability exposes them to Indirect Prompt Injection (IPI), where attackers embed malicious instructions into retrieved content to manipulate the agent into executing unauthorized or unintended actions. Existing defenses predominantly focus on the pre-processing stage, neglecting the monitoring of the model's actual behavior. In this paper, we propose PlanGuard, a training-free defense framework based on the principle of Context Isolation. Unlike prior methods, PlanGuard introduces an isolated Planner that generates a reference set of valid actions derived solely from user instructions. In addition, we design a Hierarchical Verification Mechanism that first enforces strict hard constraints to block unauthorized tool invocations, and subsequently employs an Intent Verifier to validate whether parameter deviations are benign formatting variances or malicious hijacking. Experiments on the InjecAgent benchmark demonstrate that PlanGuard effectively neutralizes these attacks, reducing the Attack Success Rate (ASR) from 72.8% to 0%, while maintaining an acceptable False Positive Rate of 1.49%. Furthermore, our method is model-agnostic and highly compatible.

Fangqi Dai, Xingjian Jiang, Zizhuang Deng

To prevent misinformation and social issues arising from trustworthy-looking content generated by LLMs, it is crucial to develop efficient and reliable methods for identifying the source of texts. Previous approaches have demonstrated exceptional performance in detecting texts fully generated by LLMs. However, these methods struggle when confronting more advanced LLM output or text with adversarial multi-task machine revision, especially in the black-box setting, where the generating model is unknown. To address this challenge, grounded in the hypothesis that human writing possesses distinctive stylistic patterns, we propose Human Language Preference Detection (HLPD). HLPD employs a reward-based alignment process, Human Language Preference Optimization (HLPO), to shift the scoring model's token distribution toward human-like writing, making the model more sensitive to human writing, therefore enhancing the identification of machine-revised text. We test HLPD in an adversarial multi-task evaluation framework that leverages a five-dimensional prompt generator and multiple advanced LLMs to create diverse revision scenarios. When detecting texts revised by GPT-series models, HLPD achieves a 15.11% relative improvement in AUROC over ImBD, surpassing Fast-DetectGPT by 45.56%. When evaluated on texts generated by advanced LLMs, HLPD achieves the highest average AUROC, exceeding ImBD by 5.53% and Fast-DetectGPT by 34.14%. Code will be made available at https://github.com/dfq2021/HLPD.

Aoying Zheng, Anqi Du, Zizhuang Deng et al.

Model quantization is a key technique for reducing storage and inference costs when deploying large language models in practice. However, recent studies show that the discretization and rounding errors introduced by quantization can be exploited by adversaries to construct quantization-conditioned backdoor (QCB) attacks. Under such attacks, malicious behaviors remain dormant in the full-precision stage and are activated only after quantized deployment, thereby bypassing conventional security auditing and detection mechanisms. To address this threat, we propose a proactive pre-quantization defense method, QuantGuard. Our method introduces differentiable rounding control variables and combines error-guided rounding reversal constraints, output-distribution consistency, and weight-distance regularization to finely regulate critical rounding behaviors. Crucially, QuantGuard utilizes only a small calibration dataset and does not modify existing quantization algorithms. This design breaks the precise alignment between attacker-crafted weight patterns and quantization boundaries, effectively suppressing the post-quantization backdoor activation pathway while preserving the model's original functionality and performance. We conduct systematic experiments on six mainstream LLMs (including the LLaMA-3 and Qwen2.5-Coder) using three quantization precisions (INT8, FP4, and NF4) across three representative scenarios: vulnerable code generation, content injection, and over-refusal. The results show that QuantGuard consistently mitigates QCB attacks, reducing the attack success rate to a level comparable to the clean model while largely preserving performance on general capability benchmarks. With low computational overhead, our method offers an effective, practical solution for secure quantized LLM deployment.

Aoying Zheng, Anqi Du, Zizhuang Deng et al.

Model quantization is essential for the efficient deployment of Large Language Models (LLMs), but introduces a critical vulnerability: Quantization-Conditioned Backdoor (QCB) attacks. In these attacks, malicious behaviors remain dormant in full-precision models and activate only after specific quantization distortions, bypassing standard security audits. To mitigate this, we introduce FlipGuard, a proactive defense framework that selectively perturbs model weights prior to quantization. By breaking the adversary's precise alignment between weight patterns and quantization boundaries, FlipGuard suppresses backdoor activation without requiring access to training data or trigger samples. We further propose the Defense Effectiveness Ratio (DER), a unified metric to jointly evaluate security gains, utility preservation, and computational cost. Extensive experiments across seven LLMs (including StarCoder and LLaMA-family models) and three quantization schemes (INT8, FP4, NF4) demonstrate that FlipGuard effectively neutralizes QCBs across three scenarios, i.e., vulnerable code generation, content injection, and over-refusal, achieving high security with negligible performance degradation.