arXiv subjects

Peiran Wang

Publications and source records attributed to Peiran Wang.

24 records · Page 2Linked to original sources

arXiv2024

RePD: Defending Jailbreak Attack through a Retrieval-based Prompt Decomposition Process

In this study, we introduce RePD, an innovative attack Retrieval-based Prompt Decomposition framework designed to mitigate the risk of jailbreak attacks on large language models (LLMs). Despite rigorous pretraining and finetuning focused on ethical alignment, LLMs are still susceptible to jailbreak exploits. RePD operates on a one-shot learning model, wherein it accesses a database of pre-collected jailbreak prompt templates to identify and decompose harmful inquiries embedded within user prompts. This process involves integrating the decomposition of the jailbreak prompt into the user's original query into a one-shot learning example to effectively teach the LLM to discern and separate malicious components. Consequently, the LLM is equipped to first neutralize any potentially harmful elements before addressing the user's prompt in a manner that aligns with its ethical guidelines. RePD is versatile and compatible with a variety of open-source LLMs acting as agents. Through comprehensive experimentation with both harmful and benign prompts, we have demonstrated the efficacy of our proposed RePD in enhancing the resilience of LLMs against jailbreak attacks, without compromising their performance in responding to typical user requests.

cs.CR↗

arXiv2024

DistDD: Distributed Data Distillation Aggregation through Gradient Matching

In this paper, we introduce DistDD, a novel approach within the federated learning framework that reduces the need for repetitive communication by distilling data directly on clients' devices. Unlike traditional federated learning that requires iterative model updates across nodes, DistDD facilitates a one-time distillation process that extracts a global distilled dataset, maintaining the privacy standards of federated learning while significantly cutting down communication costs. By leveraging the DistDD's distilled dataset, the developers of the FL can achieve just-in-time parameter tuning and neural architecture search over FL without repeating the whole FL process multiple times. We provide a detailed convergence proof of the DistDD algorithm, reinforcing its mathematical stability and reliability for practical applications. Our experiments demonstrate the effectiveness and robustness of DistDD, particularly in non-i.i.d. and mislabeled data scenarios, showcasing its potential to handle complex real-world data challenges distinctively from conventional federated learning methods. We also evaluate DistDD's application in the use case and prove its effectiveness and communication-savings in the NAS use case.

cs.LG↗

arXiv2024

Training on Fake Labels: Mitigating Label Leakage in Split Learning via Secure Dimension Transformation

Two-party split learning has emerged as a popular paradigm for vertical federated learning. To preserve the privacy of the label owner, split learning utilizes a split model, which only requires the exchange of intermediate representations (IRs) based on the inputs and gradients for each IR between two parties during the learning process. However, split learning has recently been proven to survive label inference attacks. Though several defense methods could be adopted, they either have limited defensive performance or significantly negatively impact the original mission. In this paper, we propose a novel two-party split learning method to defend against existing label inference attacks while maintaining the high utility of the learned models. Specifically, we first craft a dimension transformation module, SecDT, which could achieve bidirectional mapping between original labels and increased K-class labels to mitigate label leakage from the directional perspective. Then, a gradient normalization algorithm is designed to remove the magnitude divergence of gradients from different classes. We propose a softmax-normalized Gaussian noise to mitigate privacy leakage and make our K unknowable to adversaries. We conducted experiments on real-world datasets, including two binary-classification datasets (Avazu and Criteo) and three multi-classification datasets (MNIST, FashionMNIST, CIFAR-10); we also considered current attack schemes, including direction, norm, spectral, and model completion attacks. The detailed experiments demonstrate our proposed method's effectiveness and superiority over existing approaches. For instance, on the Avazu dataset, the attack AUC of evaluated four prominent attacks could be reduced by 0.4532+-0.0127.

cs.LG↗

arXiv2024

Moderator: Moderating Text-to-Image Diffusion Models through Fine-grained Context-based Policies

We present Moderator, a policy-based model management system that allows administrators to specify fine-grained content moderation policies and modify the weights of a text-to-image (TTI) model to make it significantly more challenging for users to produce images that violate the policies. In contrast to existing general-purpose model editing techniques, which unlearn concepts without considering the associated contexts, Moderator allows admins to specify what content should be moderated, under which context, how it should be moderated, and why moderation is necessary. Given a set of policies, Moderator first prompts the original model to generate images that need to be moderated, then uses these self-generated images to reverse fine-tune the model to compute task vectors for moderation and finally negates the original model with the task vectors to decrease its performance in generating moderated content. We evaluated Moderator with 14 participants to play the role of admins and found they could quickly learn and author policies to pass unit tests in approximately 2.29 policy iterations. Our experiment with 32 stable diffusion users suggested that Moderator can prevent 65% of users from generating moderated content under 15 attempts and require the remaining users an average of 8.3 times more attempts to generate undesired content.

cs.CR↗

arXiv2023

FedChain: An Efficient and Secure Consensus Protocol based on Proof of Useful Federated Learning for Blockchain

Blockchain has become a popular decentralized paradigm for various applications in the zero-trust environment. The core of the blockchain is the consensus protocol, which establishes consensus among all the participants. PoW (Proof-of-Work) is one of the most popular consensus protocols. However, the PoW consensus protocol which incentives the participants to use their computing power to solve a meaningless hash puzzle is continuously questioned as energy-wasting. To address these issues, we propose an efficient and secure consensus protocol based on proof of useful federated learning for blockchain (called FedChain). We first propose a secure and robust blockchain architecture that takes federated learning tasks as proof of work. Then a pool aggregation mechanism is integrated to improve the efficiency of the FedChain architecture. To protect model parameter privacy for each participant within a mining pool, a secret sharing-based ring-all reduce architecture is designed. We also introduce a data distribution-based federated learning model optimization algorithm to improve the model performance of FedChain. At last, a zero-knowledge proof-based federated learning model verification is introduced to preserve the privacy of federated learning participants while proving the model performance of federated learning participants. Our approach has been tested and validated through extensive experiments, demonstrating its performance.

cs.CR↗

arXiv2023

FedCliP: Federated Learning with Client Pruning

The prevalent communication efficient federated learning (FL) frameworks usually take advantages of model gradient compression or model distillation. However, the unbalanced local data distributions (either in quantity or quality) of participating clients, contributing non-equivalently to the global model training, still pose a big challenge to these works. In this paper, we propose FedCliP, a novel communication efficient FL framework that allows faster model training, by adaptively learning which clients should remain active for further model training and pruning those who should be inactive with less potential contributions. We also introduce an alternative optimization method with a newly defined contribution score measure to facilitate active and inactive client determination. We empirically evaluate the communication efficiency of FL frameworks with extensive experiments on three benchmark datasets under both IID and non-IID settings. Numerical results demonstrate the outperformance of the porposed FedCliP framework over state-of-the-art FL frameworks, i.e., FedCliP can save 70% of communication overhead with only 0.2% accuracy loss on MNIST datasets, and save 50% and 15% of communication overheads with less than 1% accuracy loss on FMNIST and CIFAR-10 datasets, respectively.

cs.LG↗