Metraipolis › archive › dc_6a22c8a10b34
ShadowFPT: Backdooring Federated Prompt Tuning via Shadow Triggers — arXiv:2610.22871
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.22871
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T08:42:17.790Z
- SHA-256 of the archived text
ca99f7f09f12d8182057560a108df11c3c7de4e43675550e81deaee078d6d891- Stable URL of this page
https://metraipolis.com/documento/dc_6a22c8a10b34
How to cite
ShadowFPT: Backdooring Federated Prompt Tuning via Shadow Triggers — arXiv:2610.22871 — arXiv (en), https://arxiv.org/abs/2610.22871 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6a22c8a10b34 (Metraipolis archive dc_6a22c8a10b34, sha256 ca99f7f09f12d8182057560a108df11c3c7de4e43675550e81deaee078d6d891)
BibTeX
@misc{dc6a22c8a10b34,
title = {ShadowFPT: Backdooring Federated Prompt Tuning via Shadow Triggers — arXiv:2610.22871},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.22871},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6a22c8a10b34; sha256 ca99f7f09f12d8182057560a108df11c3c7de4e43675550e81deaee078d6d891},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22871. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Federated Prompt Tuning (FPT) adapts large vision--language models by freezing the pretrained backbone and optimizing only lightweight prompt parameters across clients. Although this design improves communication and parameter efficiency, it creates an overlooked security risk. Since the backbone is shared and frozen, a malicious client can induce backdoor behavior through the representation space while keeping its uploaded prompt updates close to benign ones. We propose \textbf{ShadowFPT}, a targeted backdoor attack that exploits this frozen-backbone attack surface. ShadowFPT first pretrains a learnable \emph{Shadow Trigger} against the frozen CLIP visual encoder, using either auxiliary public data or the malicious client's local data, so that triggered inputs are steered toward the target class in representation space. During federated prompt tuning, the malicious client adapts the trigger under the current global prompt and then optimizes its local prompt on both clean and triggered samples. Only prompt parameters are uploaded to the server, while the trigger and frozen encoders remain local. By shifting most of the attack burden from prompt updates to trigger-induced representation steering, ShadowFPT achieves targeted misclassification while preserving prompt-space stealthiness. Across multiple datasets, aggregation rules, and non-IID federated partitions, ShadowFPT increases the attack success rate from 19.81\% to 90.36\% in our main setting, while maintaining clean accuracy. It remains effective across textual, visual, and joint vision--language prompt tuning. These results identify frozen backbones as stealthy and underexplored backdoor surfaces in federated prompt tuning, suggesting that defenses based only on prompt-update anomaly detection are insufficient.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.22871 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T08:39:10.040Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).