Provable one-poison backdoor attacks on linear models and ReLU neural networks

arXiv:2508.05600 · cs.LG, cs.CR · Submitted 2025-08-07 · Read on arXiv

cs.LG, cs.CR

Submitted: 2025-08-07

Updated: 2026-08-27

Comments: Added generalization to k-layer (k>=2) ReLU neural networks. Added acknowledgement

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Accuracy — 10× in this paper · explained in 1 episode
• Regularization — 10× in this paper · explained in 3 episodes
• Loss Function — 9× in this paper · explained in 2 episodes
• Ablation — 7× in this paper · explained in 1 episode
• Benchmark — 3× in this paper · explained in 2 episodes
• Generalization — 3× in this paper · explained in 13 episodes
• Adaptability — 2× in this paper · explained in 1 episode
• Generalization Bounds — 2× in this paper · explained in 2 episodes
• Feature Extraction — 1× in this paper · explained in 2 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 2 episodes

Sources

• Phantom: General Backdoor Attacks on Retrieval Augmented Language Generation
• The poison of dimensionality

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks