# DUP: Detection-guided Unlearning for Backdoor Purification in Language Models

arXiv: https://arxiv.org/abs/2508.01647 · HTML: https://mlsanity.com/paper/2508.01647/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Backdoor and data poisoning attacks, one of 12 directions in LLM safety and security.
- Builds on 19 of its 33 references that are in the map.
- First posted on arXiv 3 Aug 2025.

## Placed in

- LLM safety and security / [Backdoor and data poisoning attacks](https://mlsanity.com/safety/backdoor-and-data-poisoning-attacks/)

## Builds on (most cited first, 19 of 19)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain](https://mlsanity.com/paper/1708.06733/) (2017)
- [Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks](https://mlsanity.com/paper/1805.12185/) (2018)
- [Spectral Signatures in Backdoor Attacks](https://mlsanity.com/paper/1811.00636/) (2018)
- [Weight Poisoning Attacks on Pre-trained Models](https://mlsanity.com/paper/2004.06660/) (2020)
- [A backdoor attack against LSTM-based text classification systems](https://mlsanity.com/paper/1905.12457/) (2019)
- [BadNL: Backdoor Attacks against NLP Models with Semantic-preserving Improvements](https://mlsanity.com/paper/2006.01043/) (2020)
- [Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer](https://mlsanity.com/paper/2110.07139/) (2021)
- [Prompt as Triggers for Backdoor Attack: Examining the Vulnerability in Language Models](https://mlsanity.com/paper/2305.01219/) (2023)
- [Hidden Backdoors in Human-Centric Language Models](https://mlsanity.com/paper/2105.00164/) (2021)
- [A Unified Evaluation of Textual Backdoor Learning: Frameworks and Benchmarks](https://mlsanity.com/paper/2206.08514/) (2022)
- [Fine-mixing: Mitigating Backdoors in Fine-tuned Language Models](https://mlsanity.com/paper/2210.09545/) (2022)
- [Universal Vulnerabilities in Large Language Models: Backdoor Attacks for In-context Learning](https://mlsanity.com/paper/2401.05949/) (2024)
- [Unlearning Backdoor Attacks for LLMs with Weak-to-Strong Knowledge Distillation](https://mlsanity.com/paper/2410.14425/) (2024)
- [BadActs: A Universal Backdoor Defense in the Activation Space](https://mlsanity.com/paper/2405.11227/) (2024)
- [Expose Backdoors on the Way: A Feature-Based Efficient Defense against Textual Backdoor Attacks](https://mlsanity.com/paper/2210.07907/) (2022)
- [A Gray-box Attack against Latent Diffusion Model-based Image Editing by Posterior Collapse](https://mlsanity.com/paper/2408.10901/) (2024)

## Built on by (most cited first, 4 of 4)

- [P2P: A Poison-to-Poison Remedy for Reliable Backdoor Defense in LLMs](https://mlsanity.com/paper/2510.04503/) (2025)
- [Patronus: Identifying and Mitigating Transferable Backdoors in Pre-trained Language Models](https://mlsanity.com/paper/2512.06899/) (2025)
- [Rethinking Reasoning: A Survey on Reasoning-based Backdoors in LLMs](https://mlsanity.com/paper/2510.07697/) (2025)
- [KUDA: Knowledge Unlearning by Deviating Representation for Large Language Models](https://mlsanity.com/paper/2602.19275/) (2026)

## Abstract (arXiv)

As backdoor attacks become more stealthy and robust, they reveal critical weaknesses in current defense strategies: detection methods often rely on coarse-grained feature statistics, and purification methods typically require full retraining or additional clean models. To address these challenges, we propose DUP (Detection-guided Unlearning for Purification), a unified framework that integrates backdoor detection with unlearning-based purification. The detector captures feature-level anomalies by jointly leveraging class-agnostic distances and inter-layer transitions. These deviations are integrated through a weighted scheme to identify poisoned inputs, enabling more fine-grained analysis. Based on the detection results, we purify the model through a parameter-efficient unlearning mechanism that avoids full retraining and does not require any external clean model. Specifically, we innovatively repurpose knowledge distillation to guide the student model toward increasing its output divergence from the teacher on detected poisoned samples, effectively forcing it to unlearn the backdoor behavior. Extensive experiments across diverse attack methods and language model architectures demonstrate that DUP achieves superior defense performance in detection accuracy and purification efficacy. Our code is available at https://github.com/ManHu2025/DUP.