DYNASHIELD: A Black-Box Moving Target Defense for LLMs via Dynamic Decoding Customization
On the map: LLM safety and security → Jailbreaking
Abstract (arXiv)
Large language models (LLMs) remain vulnerable to jailbreak attacks in which adversarial prompts induce harmful outputs. Existing defenses often require access to the model internals or additional training, limiting their applicability for service providers deployed through black-box APIs. In this paper, we propose DYNASHIELD, a moving target defense framework that improves robustness by customizing decoding hyperparameters and system prompts at inference time. DYNASHIELD includes two key steps: (1) it identifies decoding configurations that reduce attack success probability, and (2) it probabilistically samples from a weighted configuration pool to introduce controlled variability in model behavior. We evaluate DYNASHIELD across 7 open-source LLMs under 4 state-of-the-art jailbreak attacks, using adversarial prompts from AdvBench. Results show substantial reductions in attack success rate compared with 7 baseline defenses, while maintaining response quality and incurring minimal inference overhead. Because DYNASHIELD operates solely through exposed runtime controls, it requires no retraining or model-internal access. These results suggest that safety-aware dynamic decoding is a promising and practically lightweight defense mechanism for black-box LLM deployments.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Jailbreaking, one of 12 directions in LLM safety and security. Builds on 28 of its 46 references that are in the map. First posted on arXiv 10 Dec 2024.
Builds on
28 of 46 references.
Closest ancestors · 11
Same direction, newest first
- 2024Probing the Safety Response Boundary of Large Language Models via Unsafe Decoding Path Generation20 Aug
- 2023DeepInception: Hypnotize Large Language Model to Be Jailbreaker6 Nov
- 2023Jailbreaking Black Box Large Language Models in Twenty Queries12 Oct
- 2023Catastrophic Jailbreak of Open-source LLMs via Exploiting Generation10 Oct
- 2023Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations10 Oct
- 2023SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks5 Oct
- 2023AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models3 Oct
- 2023GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts19 Sep
- 2023Detecting Language Model Attacks with Perplexity27 Aug
- 2023Universal and Transferable Adversarial Attacks on Aligned Language Models27 Jul
- 2023Jailbroken: How Does LLM Safety Training Fail?5 Jul
Other references · 17
- 2022Training language models to follow instructions with human feedbackcited by 15,712
- 2024The Llama 3 Herd of Modelscited by 14,192
- 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
- 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
- 2023Baseline Defenses for Adversarial Attacks Against Aligned Language Modelscited by 608
- 2023LLM Self Defense: By Self Examination, LLMs Know They Are Being Trickedcited by 226
- 2018Stochastic Activation Pruning for Robust Adversarial Defensecited by 171
- 2023Towards Efficient Generative Large Language Model Serving: A Survey from Algorithms to Systemscited by 101
- 2020SoK: Certified Robustness for Deep Neural Networkscited by 100
- 2024REVOLVE: Optimizing AI Systems by Tracking Response Evolution in Textual Optimizationcited by 20
- 2019A Research Agenda: Dynamic Models to Defend Against Correlated Attackscited by 18
- 2017MTDeep: Boosting the Security of Deep Neural Nets Against Adversarial Attacks with Moving Target Defensecited by 12
5 more
- 2023Improving the Robustness of Transformer-based Large Language Models with Dynamic Attentioncited by 10
- 2021Morphence: Moving Target Defense Against Adversarial Examplescited by 9
- 2019Moving Target Defense for Deep Visual Sensing against Adversarial Examplescited by 7
- 2019A Survey of Moving Target Defenses for Network Securitycited by 5
- 2022Morphence-2.0: Evasion-Resilient Moving Target Defense Powered by Out-of-Distribution Detectioncited by 2
Built on by
Citing papers per year: 2025: 2.
2 papers.