CodeMimicry: Exploiting Safety Generalization Lag in Large Language Models via Structured Code Completion
cs.CR, cs.AI
Submitted: 2026-09-30
Updated: 2026-09-30
Code: https://github.com/lzzzr123/CodeMimicry
Terminology
Sources
- Split and Merge: Aligning Position Biases in LLM-based Evaluators
- Improving Alignment and Robustness with Circuit Breakers
- Code Llama: Open Foundation Models for Code
- Qwen2.5-Coder Technical Report
- DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- CodeChameleon: Personalized Encryption Framework for Jailbreaking Large Language Models
- Jailbreaking Black Box Large Language Models in Twenty Queries
- GPT-4 Technical Report
- AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models
- GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher
- Jailbreaking Large Language Models with Symbolic Mathematics
- EquaCode: A Multi-Strategy Jailbreak Approach for Large Language Models via Equation Solving and Code Completion
- Tree of Attacks: Jailbreaking Black-Box LLMs Automatically
- Graph of Attacks with Pruning: Optimizing Stealthy Jailbreak Prompt Generation for Enhanced LLM Content Moderation
- Revisiting Jailbreaking for Large Language Models: A Representation Engineering Perspective
- The Internal State of an LLM Knows When It's Lying
- Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
- SelfDefend: LLMs Can Defend Themselves against Jailbreaking in a Practical Manner
- SAFEPATH: Preventing Harmful Reasoning in Chain-of-Thought via Early Alignment
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