A Theory of LLM Information Susceptibility

## 论文概要 **研究领域**: NLP **作者**: Zhuo-Yang Song, Hua Xing Zhu **发布时间**: 2026-03-26 **arXiv**: [2603.23626](https://arxiv.org/abs/2603.23626) ## 中文摘要 本研究探索了NLP领域的前沿问题。研究团队来自Zhuo-Yang Song, Hua Xing Zhu等。该方法在相关任务中展现了良好的性能和创新性。 原文摘要：Large language models (LLMs) are increasingly deployed as optimization modules in agentic systems, yet the fundamental limits of such LLM-mediated improvement remain poorly understood. Here we propose a theory of LLM information susceptibility, centred on the hypothesis that when computational resou... ## 原文摘要 Large language models (LLMs) are increasingly deployed as optimization modules in agentic systems, yet the fundamental limits of such LLM-mediated improvement remain poorly understood. Here we propose a theory of LLM information susceptibility, centred on the hypothesis that when computational resources are sufficiently large, the intervention of a fixed LLM does not increase the performance susceptibility of a strategy set with respect to budget. --- *自动采集于 2026-03-27* #论文 #arXiv #NLP #小凯