The proposed Quarantined Expert Shutdown QES is a computationally efficient containment strategy built in a regularization-steered MoE-like setting, and establishes learn, but channel as a previously unexplored regime for backdoor containment in generative LLMs.
Abstract
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.
Deep learning compilers such as TVM have become essential in deploying neural networks with high efficiency across diverse hardware. However, recent studies reveal a critical vulnerability: adversaries can embed backdoors during training and rely on compilation to re-materialize the hidden behavior in the final executa...
Ming-Hui Li, Yang Li, Li-Ming Fang· IEEE Transactions on Informa...· 0 citations
Model merging has become widely adopted for enhancing the task-specific performance of Large Language Models (LLMs) without additional training. While the specialized model from third-party platforms may undergo safety audits (e.g., backdoor detection), we reveal a novel and critical threat: One-sided Merging-Condition...
Yu-Kun Chen, Yi-Dou Chen, Yi-Ming Li et al.· 0 citations
Chorus is presented, a decentralized mechanism that lets each node detect and reject backdoored adapters from its neighbors before aggregation, without requiring shared validation data or any knowledge of the attacker's trigger or target.
Sayan Biswas, Jade Garcia Bourrée, R. Guerraoui et al.· arXiv.org· 0 citations
Large vision-language models (LVLMs) routinely use inference-time visual-token compression to prune hundreds of patch tokens down to a compact budget. Existing LVLM backdoors fire whenever a triggered image appears, whereas prior deployment-triggered attacks activate only when model weights are quantized or pruned. How...
Zhi-Fang Zhang, Qi-Rui Huang, Chen-Peng Wang et al.· 0 citations
Developers can build LLM agents by adapting third-party models through benign post-training. We study a supply-chain threat in which an attacker supplies a model with a backdoor: hidden behavior that produces malicious outputs when a particular input pattern appears. Focusing on software-engineering agents, we ask whet...
Qiusi Zhan, Nian Lyu, Stephanie Ding et al.· Conference on Empirical Meth...· 0 citations
CoDeL is presented, a defense that hardens agent against an attack distribution it reshapes as it trains, and reduces attack success rate (ASR) by 88.5% and outperforms other baselines largely (+38.0%).
Research Manager Mike Walker reflects on a career spanning hacker culture, cybersecurity, and AI, sharing lessons about trust, teamwork, leadership, and why solving difficult problems requires curiosity, judgment, agility—and an understanding of the rules. The post Learn the rules: The hacker mindset appeared first on Microsoft Research.
MIT News · Artificial Intelligence· news.mit.eduOct 13, 2026
Machine-learning networks can learn without detailed instructions from a designer. Those that do may offer insight into how the brain reorganizes itself during learning.
MIT News · Artificial Intelligence· news.mit.eduSep 29, 2026
Professor Sherry Turkle’s new book, “Artificial Intimacy,” offers a withering critique of chatbots and the antisocial dynamics she believes they encourage.