# Backdooring Bias in Large Language Models

arXiv: https://arxiv.org/abs/2602.13427 · HTML: https://mlsanity.com/paper/2602.13427/

## Key facts

- Cited by 1 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 16 of its 24 references that are in the map.
- First posted on arXiv 13 Feb 2026.

## 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, 16 of 16)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [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)
- [Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training](https://mlsanity.com/paper/2401.05566/) (2024)
- [Backdoor Learning: A Survey](https://mlsanity.com/paper/2007.08745/) (2020)
- [Backdooring Instruction-Tuned Large Language Models with Virtual Prompt Injection](https://mlsanity.com/paper/2307.16888/) (2023)
- [Composite Backdoor Attacks Against Large Language Models](https://mlsanity.com/paper/2310.07676/) (2023)
- [Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples](https://mlsanity.com/paper/2510.07192/) (2025)
- [BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models](https://mlsanity.com/paper/2406.17092/) (2024)
- [Fine-mixing: Mitigating Backdoors in Fine-tuned Language Models](https://mlsanity.com/paper/2210.09545/) (2022)
- [BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models](https://mlsanity.com/paper/2408.12798/) (2024)
- [A Survey of Recent Backdoor Attacks and Defenses in Large Language Models](https://mlsanity.com/paper/2406.06852/) (2024)
- [CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularization](https://mlsanity.com/paper/2411.12768/) (2024)
- [Augmented Neural Fine-Tuning for Efficient Backdoor Purification](https://mlsanity.com/paper/2407.10052/) (2024)

## Built on by (most cited first, 1 of 1)

- [BadDLM: Backdooring Diffusion Language Models with Diverse Targets](https://mlsanity.com/paper/2605.09397/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are increasingly deployed in settings where inducing a bias toward a certain topic can have significant consequences, and backdoor attacks can be used to produce such models. Prior work on backdoor attacks has largely focused on a black-box threat model, with an adversary targeting the model builder's LLM. However, in the bias manipulation setting, the model builder themselves could be the adversary, warranting a white-box threat model where the attacker's ability to poison, and manipulate the poisoned data is substantially increased. Furthermore, despite growing research in semantically-triggered backdoors, most studies have limited themselves to syntactically-triggered attacks. Motivated by these limitations, we conduct an analysis consisting of over 1000 evaluations using higher poisoning ratios and greater data augmentation to gain a better understanding of the potential of syntactically- and semantically-triggered backdoor attacks in a white-box setting. In addition, we study whether two representative defense paradigms, model-intrinsic and model-extrinsic backdoor removal, are able to mitigate these attacks. Our analysis reveals numerous new findings. We discover that while both syntactically- and semantically-triggered attacks can effectively induce the target behaviour, and largely preserve utility, semantically-triggered attacks are generally more effective in inducing negative biases, while both backdoor types struggle with causing positive biases. Furthermore, while both defense types are able to mitigate these backdoors, they either result in a substantial drop in utility, or require high computational overhead.