Searcharxiv⌕ Search

arXiv subjects

Wuqiong Pan

Publications and source records attributed to Wuqiong Pan.

3 recordsLinked to original sources

MGE: A Membership Guided Data Extraction Attack against Diffusion Models

Recently, diffusion models have achieved remarkable performance in image generation. However, their inherent memory properties introduce a potential safety risk of training data leakage, i.e., data extraction attack. To illustrate such risk, we propose a membership guided extraction (MGE) framework to effectively extract training data from a pre-trained diffusion model. Our core idea is to formulate the data extraction as a conditional generation, where the condition signal is from a membership inference classifier. To improve the success rate and efficiency of MGE, we propose adversarial gradient and parallel branch decoding. Our experiments on image generation benchmarks demonstrate that our MGE can extract hundreds of images, which shows that more attention should be paid to privacy-preserving of diffusion model.

cs.CV↗

Finding Where the Buck Stops: An Automated Failure Attribution-Based Reflection Framework for Multi-Agent Collaboration

Multi-agent systems (MAS) powered by large language models have shown promise for complex tasks but suffer from high failure rates. Current self-reflection methods for MAS require all agents to reflect upon failure, overlooking a critical reality: failures typically stem from a specific agent leading the task astray, namely the decisive error agent, while others merely fulfill their regular duties. Forcing regular-behaving agents to reflect contaminates their memory with wrong insights. Hence, we propose DoCtOR (Diagnose-then-Correct PPO-enhanced Reflection), a novel reflection framework that enhances multi-agent collaboration. DoCtOR first identifies the decisive error step and decisive error agent through automated failure attribution, then employs counterfactual reasoning to generate a corrected decisive error step, and finally engages only the decisive error agent to produce targeted reflections. Experimental results show DoCtOR achieves 22%, 26%, and 27% improvements over initial success rates on HotPotQA, ChartQAPro, and Mind2Web datasets, outperforming Reflexion, Retroformer, and COPPER. We further establish the generalizability of our diagnose-then-correct paradigm and demonstrate that in low-resource settings, focusing reflection on reasoning steps after the decisive error step achieves comparable quality to reflecting on the complete failure trajectory.

cs.AI↗

A new lightweight additive homomorphic encryption algorithm

This article describes a lightweight additive homomorphic algorithm with the same encryption and decryption keys. Compared to standard additive homomorphic algorithms like Paillier, this algorithm reduces the computational cost of encryption and decryption from modular exponentiation to modular multiplication, and reduces the computational cost of ciphertext addition from modular multiplication to modular addition. This algorithm is based on a new mathematical problem: in two division operations, whether it is possible to infer the remainder or divisor based on the dividend when two remainders are related. Currently, it is not obvious how to break this problem, but further exploration is needed to determine if it is sufficiently difficult. In addition to this mathematical problem, we have also designed two interesting mathematical structures for decryption, which are used in the two algorithms mentioned in the main text. It is possible that the decryption structure of Algorithm 2 introduces new security vulnerabilities, but we have not investigated this issue thoroughly.

cs.CR↗