Prompt Specificity in River-Crossing Logic: A Comparative Study of General-Purpose vs. Reasoning-Optimized LLMs

This paper investigates how different prompting strategies affect large language models (LLMs) 
when solving a logic puzzle. We compare “general-purpose” models vs. “reasoning-optimized” 
models under various prompt conditions, providing insights into prompt engineering best practices. 
Licensed under CC-BY 4.0.