Impact of Syntactic Perturbation in Arabic Self-Invoking Code on Multilingual LLM Pass@k Metrics Relative to English Baselines

In an era dominated by Large Language Models (LLMs), understanding their capabilities and limitations, especially in high-stakes fields like law, is crucial. While LLMs such as Meta's LLaMA, OpenAI's ChatGPT, Google's Gemini, DeepSeek, and other emerging models are increasingly integrated into legal workflows, their performance in multilingual, jurisdictionally diverse, and adversarial contexts remains insufficiently explored. This work evaluates LLaMA and Gemini on multilingual legal and non-legal benchmarks, and assesses their adversarial robustness in legal tasks through character and word-

Research goal: How does syntactic perturbation in Arabic self-invoking code tasks impact pass@k metrics of multilingual LLMs relative to English baselines on HumanEval Pro?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 7.6/10.