Model Releases
Large-scale Testing Global Optimization Methods with Black-box Adversarial Attacks
arXiv:2608.13296v1 Announce Type: cross Abstract: Existing global optimization benchmark suites are of a moderate size and are based on a small number of analytical functions that date back even to th
arXiv:2608.13296v1 Announce Type: cross Abstract: Existing global optimization benchmark suites are of a moderate size and are based on a small number of analytical functions that date back even to the 1970s. This causes a risk of biasing the development of global optimization methods. We argue that the tasks related to the black-box adversarial attack (BBAA) can serve as valuable global optimization benchmark in many-dimensional space. We demonstrate the efficiency of several types of evolutionary algorithms and other metaheuristics in solving example BBAA problems. Thus, we take a step towards convergence of global optimization methods to the challenges and needs that arise in the modern machine learning field.
Related
- Babel: Jailbreaking Safety Attention via Obfuscation Distribution Optimized Sampling
- Metis: Learning to Jailbreak LLMs via Self-Evolving Metacognitive Policy Optimization
- Global optimization tailored for graphics processing units: Complete and rigorous search for large-scale nonlinear minimization
- Dynamic Jailbreaking Attack
- Detecting Fluent Optimization-Based Adversarial Prompts via Sequential Entropy Changes
Source: arXiv cs.AI | 2026-08-14