Cover Image

Circumventing Backdoor Space via Weight Symmetry

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Peng, Jie, Yang, Hongwei, Zhao, Jing, Dong, Hengji, He, Hui, Zhang, Weizhe, He, Haoyu
Format:	Preprint
Published:	2025
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Preserving Continuous Symmetry in Discrete Spaces: Geometric-Aware Quantization for SO(3)-Equivariant GNNs
by: Zhou, Haoyu, et al.
Published: (2026)

Understanding Mode Connectivity via Parameter Space Symmetry
by: Zhao, Bo, et al.
Published: (2025)

Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness
by: Lin, Weilin, et al.
Published: (2024)

Instruction Backdoor Attacks Against Customized LLMs
by: Zhang, Rui, et al.
Published: (2024)

Watch Out for Your Guidance on Generation! Exploring Conditional Backdoor Attacks against Large Language Models
by: He, Jiaming, et al.
Published: (2024)

Exploit Gradient Skewness to Circumvent Byzantine Defenses for Federated Learning
by: Liu, Yuchen, et al.
Published: (2025)

Latent Space Symmetry Discovery
by: Yang, Jianke, et al.
Published: (2023)

Membership Inference Attack with Partial Features
by: Wang, Xurun, et al.
Published: (2025)

Revisiting Weighted Strategy for Non-stationary Parametric Bandits and MDPs
by: Wang, Jing, et al.
Published: (2026)

Trusted Weights, Treacherous Optimizations? Optimization-Triggered Backdoor Attacks on LLMs
by: Wang, Yifei, et al.
Published: (2026)

Poisoning with A Pill: Circumventing Detection in Federated Learning
by: Guo, Hanxi, et al.
Published: (2024)

FLAIN: Mitigating Backdoor Attacks in Federated Learning via Flipping Weight Updates of Low-Activation Input Neurons
by: Ding, Binbin, et al.
Published: (2024)

InkDrop: Invisible Backdoor Attacks Against Dataset Condensation
by: Yang, He, et al.
Published: (2026)

Magnitude-based Neuron Pruning for Backdoor Defens
by: Li, Nan, et al.
Published: (2024)

Authority Backdoor: A Certifiable Backdoor Mechanism for Authoring DNNs
by: Yang, Han, et al.
Published: (2025)

State Backdoor: Towards Stealthy Real-world Poisoning Attack on Vision-Language-Action Model in State Space
by: Guo, Ji, et al.
Published: (2026)

Sample Weight Averaging for Stable Prediction
by: Yu, Han, et al.
Published: (2025)

Backdoor for Debias: Mitigating Model Bias with Backdoor Attack-based Artificial Bias
by: Wu, Shangxi, et al.
Published: (2023)

Effective Sparsity: A Unified Framework via Normalized Entropy and the Effective Number of Nonzeros
by: He, Haoyu, et al.
Published: (2026)

Symmetry in Neural Network Parameter Spaces
by: Zhao, Bo, et al.
Published: (2025)

MemoryGraft: Persistent Compromise of LLM Agents via Poisoned Experience Retrieval
by: Srivastava, Saksham Sahai, et al.
Published: (2025)

Certifying Language Model Robustness with Fuzzed Randomized Smoothing: An Efficient Defense Against Backdoor Attacks
by: He, Bowei, et al.
Published: (2025)

A Semantic and Clean-label Backdoor Attack against Graph Convolutional Networks
by: Dai, Jiazhu, et al.
Published: (2025)

Improved Generalization of Weight Space Networks via Augmentations
by: Shamsian, Aviv, et al.
Published: (2024)

Turn-Based Structural Triggers: Prompt-Free Backdoors in Multi-Turn LLMs
by: Lu, Yiyang, et al.
Published: (2026)

ArcGen: Generalizing Neural Backdoor Detection Across Diverse Architectures
by: Yang, Zhonghao, et al.
Published: (2025)

Weight Space Representation Learning via Neural Field Adaptation
by: Yang, Zhuoqian, et al.
Published: (2025)

Vertical Federated Unlearning via Backdoor Certification
by: Han, Mengde, et al.
Published: (2024)

Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses
by: Zheng, Xiaosen, et al.
Published: (2024)

Beyond Uniform Deletion: A Data Value-Weighted Framework for Certified Machine Unlearning
by: He, Lisong, et al.
Published: (2025)

Improving the Sensitivity of Backdoor Detectors via Class Subspace Orthogonalization
by: Yang, Guangmingmei, et al.
Published: (2025)

Your Compiler is Backdooring Your Model: Understanding and Exploiting Compilation Inconsistency Vulnerabilities in Deep Learning Compilers
by: Chen, Simin, et al.
Published: (2025)

FedBAP: Backdoor Defense via Benign Adversarial Perturbation in Federated Learning
by: Yan, Xinhai, et al.
Published: (2025)

Fusing Pruned and Backdoored Models: Optimal Transport-based Data-free Backdoor Mitigation
by: Lin, Weilin, et al.
Published: (2024)

Out-of-Support Generalisation via Weight-Space Sequence Modelling
by: Nzoyem, Roussel Desmond
Published: (2026)

CS-GBA: A Critical Sample-based Gradient-guided Backdoor Attack for Offline Reinforcement Learning
by: Zhao, Yuanjie, et al.
Published: (2026)

FAAST: Forward-Only Associative Learning via Closed-Form Fast Weights for Test-Time Supervised Adaptation
by: Bao, Guangsheng, et al.
Published: (2026)

Non-stationary Online Learning for Curved Losses: Improved Dynamic Regret via Mixability
by: Zhang, Yu-Jie, et al.
Published: (2025)

BAPFL: Exploring Backdoor Attacks Against Prototype-based Federated Learning
by: Zeng, Honghong, et al.
Published: (2025)

FedAWA: Adaptive Optimization of Aggregation Weights in Federated Learning Using Client Vectors
by: Shi, Changlong, et al.
Published: (2025)