mlsanityData to 14 Oct 2026

ADCA: Attention-Driven Multi-Party Collusion Attack in Federated Self-Supervised Learning

arXiv 2602.05612 ↗First posted 5 Feb 2026cs.CRcited by 0

On the map: LLM safety and security → Federated unlearning and poisoning

Abstract (arXiv)

Federated Self-Supervised Learning (FSSL) integrates the privacy advantages of distributed training with the capability of self-supervised learning to leverage unlabeled data, showing strong potential across applications. However, recent studies have shown that FSSL is also vulnerable to backdoor attacks. Existing attacks are limited by their trigger design, which typically employs a global, uniform trigger that is easily detected, gets diluted during aggregation, and lacks robustness in heterogeneous client environments. To address these challenges, we propose the Attention-Driven multi-party Collusion Attack (ADCA). During local pre-training, malicious clients decompose the global trigger to find optimal local patterns. Subsequently, these malicious clients collude to form a malicious coalition and establish a collaborative optimization mechanism within it. In this mechanism, each submits its model updates, and an attention mechanism dynamically aggregates them to explore the best cooperative strategy. The resulting aggregated parameters serve as the initial state for the next round of training within the coalition, thereby effectively mitigating the dilution of backdoor information by benign updates. Experiments on multiple FSSL scenarios and four datasets show that ADCA significantly outperforms existing methods in Attack Success Rate (ASR) and persistence, proving its effectiveness and robustness.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Federated unlearning and poisoning, one of 12 directions in LLM safety and security. Builds on 23 of its 39 references that are in the map. First posted on arXiv 5 Feb 2026.

Builds on

23 of 39 references.

References · 23

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  4. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  5. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,513
  6. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,750
  7. 2020Exploring Simple Siamese Representation Learningcited by 2,866
  8. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,857
  9. 2018How To Backdoor Federated Learningcited by 765
  10. 2021Data-Free Knowledge Distillation for Heterogeneous Federated Learningcited by 383
  11. 2020FLTrust: Byzantine-robust Federated Learning via Trust Bootstrappingcited by 346
  12. 2019Can You Really Backdoor Federated Learning?cited by 255
11 more
  1. 2021BadEncoder: Backdoor Attacks to Pre-trained Encoders in Self-Supervised Learningcited by 123
  2. 2021Backdoor Attacks on Self-Supervised Learningcited by 90
  3. 2022Divergence-aware Federated Self-Supervised Learningcited by 73
  4. 2022An Embarrassingly Simple Backdoor Attack on Self-supervised Learningcited by 31
  5. 2023L-DAWA: Layer-wise Divergence Aware Weight Aggregation in Federated Self-Supervised Visual Representation Learningcited by 30
  6. 2021SSFL: Tackling Label Deficiency in Federated Learning via Personalized Self-Supervisioncited by 27
  7. 2023Defending Against Patch-based Backdoor Attacks on Self-Supervised Learningcited by 20
  8. 2024Beyond Traditional Threats: A Persistent Backdoor Attack on Federated Learningcited by 16
  9. 2023ProtoFL: Unsupervised Federated Learning via Prototypical Distillationcited by 12
  10. 2023Erasing Self-Supervised Learning Backdoor by Cluster Activation Maskingcited by 2
  11. 2024EmInspector: Combating Backdoor Attacks in Federated Self-Supervised Learning Through Embedding Inspectioncited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Non-Cooperative Backdoor Attacks in Federated Learning: A New Threat Landscape
  2. 2025IPBA: Imperceptible Perturbation Backdoor Attack in Federated Self-Supervised Learning
  3. 2025Enhancing the Effectiveness and Durability of Backdoor Attacks in Federated Learning through Maximizing Task Distinction
  4. 2025DOPA: Stealthy and Generalizable Backdoor Attacks from a Single Client under Challenging Federated Constraints ↗
  5. 2025FedBAP: Backdoor Defense via Benign Adversarial Perturbation in Federated Learning
  6. 2025A Client-level Assessment of Collaborative Backdoor Poisoning in Non-IID Federated Learning
  7. 2026Mitigating Backdoor Attacks in Federated Learning Using PPA and MiniMax Game Theory
  8. 2023BAGEL: Backdoor Attacks against Federated Contrastive Learning