Metraipolis › archive › dc_8f2bdd6615fe

Backdoor Containment via Expert Quarantine and Shutdown in LLMs — arXiv:2610.00663

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.00663
License (as archived)
CC0-1.0
Archived
2026-10-06T12:02:31.447Z
SHA-256 of the archived text
7aa19f9b1012fa2c46e8432939edc8e00ee5ce122ddbb2c103593b0a4127b608
Stable URL of this page
https://metraipolis.com/documento/dc_8f2bdd6615fe

How to cite

Backdoor Containment via Expert Quarantine and Shutdown in LLMs — arXiv:2610.00663 — arXiv (en), https://arxiv.org/abs/2610.00663 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8f2bdd6615fe (Metraipolis archive dc_8f2bdd6615fe, sha256 7aa19f9b1012fa2c46e8432939edc8e00ee5ce122ddbb2c103593b0a4127b608)
BibTeX
@misc{dc8f2bdd6615fe,
  title = {Backdoor Containment via Expert Quarantine and Shutdown in LLMs — arXiv:2610.00663},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.00663},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8f2bdd6615fe; sha256 7aa19f9b1012fa2c46e8432939edc8e00ee5ce122ddbb2c103593b0a4127b608},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00663. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Backdoored large language models (LLMs) can behave normally on benign inputs while producing attacker-specified outputs under hidden triggers. Existing defenses span four stages--prior-training, in-training, post-training, and inference-time--and share one of two underlying strategies: either suppress backdoor learning (by filtering poisoned data or interrupting its acquisition during optimization) or learn, then purify (by repairing model weights or gating inputs after a fully backdoored model has formed). We propose a third strategy, learn, but channel: allow backdoor formation during training but route it into a designated, quarantined component that can be disabled at deployment. To this end, we propose Quarantined Expert Shutdown QES, a computationally efficient containment strategy built in a regularization-steered MoE-like setting. Specifically, given a poisoned dataset, QES augments a Transformer-based language model with routed expert-specific LoRA branches and lightweight routers, and uses auxiliary routing objectives to attract trigger-conditioned behavior into a designated expert while preserving benign capability elsewhere. At deployment, mitigation reduces to a single constant-time operation: zeroing the quarantined expert's routing weight, without trigger screening or further updating model weights. Empirically, our methods reduce the attack success rate ASR from 100% to 0-10% on most settings across two tasks, three attacks, and four model families, while downstream utility is often preserved or only modestly affected. These results establish learn, but channel as a previously unexplored regime for backdoor containment in generative LLMs.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Cryptography and Security (cs.CR); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.00663 — 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 2026-10-06T10:21:27.879Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).