![[2511.14406] Watch Out for the Lifespan: Evaluating Backdoor Attacks Against Federated Model Adaptation](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2511.14406] Watch Out for the Lifespan: Evaluating Backdoor Attacks Against Federated Model Adaptation
Summary
This paper evaluates backdoor attacks against federated learning model adaptation, focusing on the impact of Low-Rank Adaptation (LoRA) on attack persistence and security vulnerabilities.
Why It Matters
As federated learning becomes more prevalent in machine learning applications, understanding the security risks, particularly backdoor attacks, is crucial. This research sheds light on how model adaptation techniques like LoRA can influence the effectiveness and longevity of such attacks, informing better security practices in federated systems.
Key Takeaways
- LoRA can affect the persistence of backdoor attacks in federated learning.
- Lower ranks in LoRA lead to longer backdoor lifespan post-attack.
- The study addresses critical evaluation issues in assessing backdoor attacks.
- Improving understanding of backdoor risks enhances federated learning reliability.
- The research contributes to developing robust evaluation frameworks for security in FL.
Computer Science > Machine Learning arXiv:2511.14406 (cs) [Submitted on 18 Nov 2025 (v1), last revised 18 Feb 2026 (this version, v2)] Title:Watch Out for the Lifespan: Evaluating Backdoor Attacks Against Federated Model Adaptation Authors:Bastien Vuillod, Pierre-Alain Moellic, Jean-Max Dutertre View a PDF of the paper titled Watch Out for the Lifespan: Evaluating Backdoor Attacks Against Federated Model Adaptation, by Bastien Vuillod and 2 other authors View PDF HTML (experimental) Abstract:Large models adaptation through Federated Learning (FL) addresses a wide range of use cases and is enabled by Parameter-Efficient Fine-Tuning techniques such as Low-Rank Adaptation (LoRA). However, this distributed learning paradigm faces several security threats, particularly to its integrity, such as backdoor attacks that aim to inject malicious behavior during the local training steps of certain clients. We present the first analysis of the influence of LoRA on state-of-the-art backdoor attacks targeting model adaptation in FL. Specifically, we focus on backdoor lifespan, a critical characteristic in FL, that can vary depending on the attack scenario and the attacker's ability to effectively inject the backdoor. A key finding in our experiments is that for an optimally injected backdoor, the backdoor persistence after the attack is longer when the LoRA's rank is lower. Importantly, our work highlights evaluation issues of backdoor attacks against FL and contributes to the developmen...
