LLM-SPICEMixer improves median final training reward by 8.4% and median validation-selected test reward by 8.8% compared to genetic framework without LLM guidance

Compared with the genetic framework without LLM guidance, LLM-SPICEMixer improves the median final training reward by 8.4% and the median validation-selected test reward by 8.8%

http://arxiv.org/abs/2608.23317v1