EUVIMEDEUVIMED entdecken
Crossref

Scaling the Strategy Wall: Efficient Jailbreaking of LLMs via Component-Based Multi-Objective Optimization

Jialing Tao, Song Huang, Changyou Zheng

Computers, Materials & Continua · 2026 · 88 (3) · 1-10

DOI: 10.32604/cmc.2026.080119 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computers, Materials & Continua, 2026. Verfasst von Jialing Tao und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

: Background: Jailbreak attacks, which use crafted prompts to bypass safety alignments of Large Language Models (LLMs) and generate harmful content, pose a significant security threat. Existing methods often optimize for a single objective (e.g., attack success rate), neglecting critical factors like query efficiency, which limits their practicality and generalization.

Methods

We propose a Componentized Multi-Objective Optimization Framework (CMOOF), which introduces a paradigm shift: it searches for generalizable and query-efficient attack strategy templates within a structured, component-based strategy space. CMOOF leverages the NSGA-II algorithm to explicitly co-optimize two first-class objectives: Attack Success Rate (ASR) and Query Efficiency, thereby discovering their Pareto-optimal trade-off frontier.

Results

Experiments on benchmark datasets show significant improvements, with the highest jailbreak success rate reaching 98.75% on models like Llama3, and query efficiency surpassing baselines.

Conclusions

CMOOF redefines jailbreak optimization from instance-level prompt crafting to strategy-level template discovery. The work provides an efficient, scalable, and generalizable jailbreak solution, and the framework offers broader insights for automated red teaming and LLM security defense.

Zitieren

APA

Tao, J., Huang, S., & Zheng, C. (2026). Scaling the Strategy Wall: Efficient Jailbreaking of LLMs via Component-Based Multi-Objective Optimization. Computers, Materials & Continua, 88(3), 1-10. https://doi.org/10.32604/cmc.2026.080119

Vancouver

Tao J, Huang S, Zheng C. Scaling the Strategy Wall: Efficient Jailbreaking of LLMs via Component-Based Multi-Objective Optimization. Computers, Materials & Continua. 2026;88(3):1-10. doi:10.32604/cmc.2026.080119

BibTeX
@article{Tao2026,
  author = {Tao, Jialing and Huang, Song and Zheng, Changyou},
  title = {Scaling the Strategy Wall: Efficient Jailbreaking of LLMs via Component-Based Multi-Objective Optimization},
  journal = {Computers, Materials & Continua},
  year = {2026},
  volume = {88},
  number = {3},
  pages = {1--10},
  doi = {10.32604/cmc.2026.080119}
}
Kostenlos starten →