Harnessing non-adversarial robustness in large language models 事件

Harnessing non-adversarial robustness in large language models arXiv:2605.29816v1 Announce Type: new Abstract: The work presents an approach for addressing the challenge of robustness in Large Language Models (LLMs) to alterations and potential errors caused by semantically similar but textually different prompts. Recent works have shown that these kinds of prompt variations can significantly impact the performance of LLMs on tasks. The central question is: can LLMs' robustness to semantically-