Securing LLMs in the Wild: Privacy and Security Challenges at the Edge
Ren-Yi Huang, Mingchen Li, Dumindu Samaraweera, Morris Chang
cs.CR, cs.LG
Submitted: 2026-07-13
Code: https://github.com/hendrycks/test
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Advancing Practical Homomorphic Encryption for Federated Learning: Theoretical Guarantees and Efficiency Optimizations
- Explaining and Harnessing Adversarial Examples
- Attacking Binarized Neural Networks
- The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks
- Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications
- Palu: Compressing KV-Cache with Low-Rank Projection
- LoRA as Oracle
- AlignGuard-LoRA: Alignment-Preserving Fine-Tuning via Fisher-Guided Decomposition and Riemannian-Geodesic Collision Regularization
- Analyzing and Reducing Catastrophic Forgetting in Parameter Efficient Tuning
- Measuring Massive Multitask Language Understanding
- Evaluating Large Language Models Trained on Code
- Phi-3 Safety Post-Training: Aligning Language Models with a "Break-Fix" Cycle
- Qwen2.5-Coder Technical Report
- Gemma 2: Improving Open Language Models at a Practical Size
- The Llama 3 Herd of Models
- Decoding Compressed Trust: Scrutinizing the Trustworthiness of Efficient LLMs Under Compression
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills
- Token Composition: A Graph Based on EVM Logs