Evolutionary computations (as innovative mathematical programming branch) in crisis-resilient computational linguistics

Abstract

The paper explores evolutionary computing as an innovative branch of mathematical programming applied to modern computational linguistics under crisis and uncertainty. The study emphasizes the role of evolutionary algorithms–such as genetic algorithms, evolutionary strategies, and genetic programming–in developing adaptive, flexible, and resilient linguistic systems. Unlike traditional gradient-based optimization methods, evolutionary computation enables a global search for optimal solutions in noisy, incomplete, and dynamically changing linguistic environments. The integration of these methods with large language models (LLMs) enhances their adaptability, reliability, and efficiency in processing crisis-related or resource-scarce data. As a result, evolutionary computing provides the foundation for hybrid architectures capable of maintaining performance stability and linguistic intelligence in complex and volatile contexts.