# Mitigating Fine-tuning based Jailbreak Attack with Backdoor Enhanced Safety Alignment

arXiv: https://arxiv.org/abs/2402.14968 · HTML: https://mlsanity.com/paper/2402.14968/

## Key facts

- Cited by 31 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 44 references that are in the map.
- First posted on arXiv 22 Feb 2024.

## Builds on (most cited first, 30 of 34)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [Sparks of Artificial General Intelligence: Early experiments with GPT-4](https://mlsanity.com/paper/2303.12712/) (2023)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding](https://mlsanity.com/paper/2306.02858/) (2023)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning](https://mlsanity.com/paper/1712.05526/) (2017)
- [LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention](https://mlsanity.com/paper/2303.16199/) (2023)
- [Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning](https://mlsanity.com/paper/1709.00103/) (2017)
- [Instruction Tuning for Large Language Models: A Survey](https://mlsanity.com/paper/2308.10792/) (2023)
- [SAMSum Corpus: A Human-annotated Dialogue Dataset for Abstractive Summarization](https://mlsanity.com/paper/1911.12237/) (2019)
- [Backdoor Learning: A Survey](https://mlsanity.com/paper/2007.08745/) (2020)
- [Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networks](https://mlsanity.com/paper/2007.02343/) (2020)
- [AgentTuning: Enabling Generalized Agent Abilities for LLMs](https://mlsanity.com/paper/2310.12823/) (2023)
- [Shadow Alignment: The Ease of Subverting Safely-Aligned Language Models](https://mlsanity.com/paper/2310.02949/) (2023)
- [A backdoor attack against LSTM-based text classification systems](https://mlsanity.com/paper/1905.12457/) (2019)
- [LoRA Fine-tuning Efficiently Undoes Safety Training in Llama 2-Chat 70B](https://mlsanity.com/paper/2310.20624/) (2023)
- [Removing RLHF Protections in GPT-4 via Fine-Tuning](https://mlsanity.com/paper/2311.05553/) (2023)
- [Universal Jailbreak Backdoors from Poisoned Human Feedback](https://mlsanity.com/paper/2311.14455/) (2023)
- [Instructions as Backdoors: Backdoor Vulnerabilities of Instruction Tuning for Large Language Models](https://mlsanity.com/paper/2305.14710/) (2023)

## Built on by (most cited first, 30 of 31)

- [Safe LoRA: the Silver Lining of Reducing Safety Risks when Fine-tuning Large Language Models](https://mlsanity.com/paper/2405.16833/) (2024)
- [Process Reward Model with Q-Value Rankings](https://mlsanity.com/paper/2410.11287/) (2024)
- [Foot-In-The-Door: A Multi-turn Jailbreak for LLMs](https://mlsanity.com/paper/2502.19820/) (2025)
- [JailbreakEval: An Integrated Toolkit for Evaluating Jailbreak Attempts Against Large Language Models](https://mlsanity.com/paper/2406.09321/) (2024)
- [No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks](https://mlsanity.com/paper/2405.16229/) (2024)
- [Adversarial Tuning: Defending Against Jailbreak Attacks for LLMs](https://mlsanity.com/paper/2406.06622/) (2024)
- [SafeMERGE: Preserving Safety Alignment in Fine-Tuned Large Language Models via Selective Layer-Wise Model Merging](https://mlsanity.com/paper/2503.17239/) (2025)
- [Why LLM Safety Guardrails Collapse After Fine-tuning: A Similarity Analysis Between Alignment and Fine-tuning Datasets](https://mlsanity.com/paper/2506.05346/) (2025)
- [LoX: Low-Rank Extrapolation Robustifies LLM Safety Against Fine-tuning](https://mlsanity.com/paper/2506.15606/) (2025)
- [Adaptive Defense against Harmful Fine-Tuning for Large Language Models via Bayesian Data Scheduler](https://mlsanity.com/paper/2510.27172/) (2025)
- [Towards Safeguarding LLM Fine-tuning APIs against Cipher Attacks](https://mlsanity.com/paper/2508.17158/) (2025)
- [Fundamental Limitations in Pointwise Defences of LLM Finetuning APIs](https://mlsanity.com/paper/2502.14828/) (2025)
- [On the Learnability of Out-of-distribution Detection](https://mlsanity.com/paper/2404.04865/) (2024)
- [Why Not Act on What You Know? Unleashing Safety Potential of LLMs via Self-Aware Guard Enhancement](https://mlsanity.com/paper/2505.12060/) (2025)
- [What Really Matters in Many-Shot Attacks? An Empirical Study of Long-Context Vulnerabilities in LLMs](https://mlsanity.com/paper/2505.19773/) (2025)
- [No, of Course I Can! Deeper Fine-Tuning Attacks That Bypass Token-Level Safety Mechanisms](https://mlsanity.com/paper/2502.19537/) (2025)
- [Layer-Aware Task Arithmetic: Disentangling Task-Specific and Instruction-Following Knowledge](https://mlsanity.com/paper/2502.20186/) (2025)
- [Safety Subspaces are Not Linearly Distinct: A Fine-Tuning Case Study](https://mlsanity.com/paper/2505.14185/) (2025)
- [AdvBDGen: Adversarially Fortified Prompt-Specific Fuzzy Backdoor Generator Against LLM Alignment](https://mlsanity.com/paper/2410.11283/) (2024)
- [SC-LoRA: Balancing Efficient Fine-tuning and Knowledge Preservation via Subspace-Constrained LoRA](https://mlsanity.com/paper/2505.23724/) (2025)
- [Few Tokens, Big Leverage: Preserving Safety Alignment by Constraining Safety Tokens during Fine-tuning](https://mlsanity.com/paper/2603.07445/) (2026)
- [Detecting Adversarial Fine-tuning with Auditing Agents](https://mlsanity.com/paper/2510.16255/) (2025)
- [Attention-Aware GNN-based Input Defense against Multi-Turn LLM Jailbreak](https://mlsanity.com/paper/2507.07146/) (2025)
- [Safeguarding LLM Fine-tuning via Push-Pull Distributional Alignment](https://mlsanity.com/paper/2601.07200/) (2026)
- [TamperBench: Systematically Stress-Testing LLM Safety Under Fine-Tuning and Tampering](https://mlsanity.com/paper/2602.06911/) (2026)
- [THRD: A Training-Free Multi-Turn Defense Framework for Jailbreak Attacks on Large Language Models](https://mlsanity.com/paper/2606.01738/) (2026)
- [SafeGene: Reusable Adapters for Transferable Safety Alignment](https://mlsanity.com/paper/2606.06519/) (2026)
- [Multilingual Collaborative Defense for Large Language Models](https://mlsanity.com/paper/2505.11835/) (2025)
- [Can LLM Safety Be Ensured by Constraining Parameter Regions?](https://mlsanity.com/paper/2602.17696/) (2026)
- [Position: Retire the "Positive Backdoor" Label -- Secret Alignment Requires Strict and Systematic Evaluation](https://mlsanity.com/paper/2605.28597/) (2026)

## Abstract (arXiv)

Despite the general capabilities of Large Language Models (LLM), these models still request fine-tuning or adaptation with customized data when meeting specific business demands. However, this process inevitably introduces new threats, particularly against the Fine-tuning based Jailbreak Attack (FJAttack) under the setting of Language-Model-as-a-Service (LMaaS), where the model's safety has been significantly compromised by fine-tuning users' uploaded examples contain just a few harmful examples. Though potential defenses have been proposed that the service providers can integrate safety examples into the fine-tuning dataset to reduce safety issues, such approaches require incorporating a substantial amount of data, making it inefficient. To effectively defend against the FJAttack with limited safety examples under LMaaS, we propose the Backdoor Enhanced Safety Alignment method inspired by an analogy with the concept of backdoor attacks. In particular, service providers will construct prefixed safety examples with a secret prompt, acting as a "backdoor trigger". By integrating prefixed safety examples into the fine-tuning dataset, the subsequent fine-tuning process effectively acts as the "backdoor attack", establishing a strong correlation between the secret prompt and safety generations. Consequently, safe responses are ensured once service providers prepend this secret prompt ahead of any user input during inference. Our comprehensive experiments demonstrate that through the Backdoor Enhanced Safety Alignment with adding as few as 11 prefixed safety examples, the maliciously fine-tuned LLMs will achieve similar safety performance as the original aligned models without harming the benign performance. Furthermore, we also present the effectiveness of our method in a more practical setting where the fine-tuning data consists of both FJAttack examples and the fine-tuning task data.