AI RESEARCH
TextReg: Mitigating Prompt Distributional Overfitting via Regularized Text-Space Optimization
arXiv CS.LG
•
ArXi:2605.21318v1 Announce Type: cross Large language models (LLMs) are highly sensitive to the prompts used to specify task objectives and behavioral constraints. Many recent prompt optimization methods iteratively rewrite prompts using LLM-generated feedback, but the resulting prompts often become longer, accumulate narrow sample-specific rules, and generalize poorly beyond the