Connecting Large Language Models with Evolutionary Algorithms Yields Powerful Prompt Optimizers
By Qingyan Guo, Rui Wang, Junliang Guo, Bei Li, Kaitao Song, Xu Tan, Guoqing Liu, Jiang Bian, Yujiu Yang. International Conference on Learning Representations, 2024
334 citations recorded by the source corpus.
Subfields: Artificial Intelligence
View Paper: https://www.semanticscholar.org/paper/8d17234680db76f99efd22fbcb169f45d2d79d93
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Osbert Bastani
- Anima Anandkumar
- Jiarui Wang
- Yecheng Jason Ma
- Haoran Ye
- Zhiguang Cao
- Guojie Song
- Guanzhi Wang
Citations in the corpus
- Eureka: Human-Level Reward Design via Coding Large Language Models (2023, positive): “Evolution with LLMs. Implementing evolutionary algorithms with LLMs has been explored in recent works in the context of neural architecture search (Chen et al., 2023; Nasir et al., 2023), prompt engineering (Guo et al., 2023), as well as morphology design (Lehman et al., 2022).”
- Evaluation and mitigation of the limitations of large language models in clinical decision-making (2024, neutral): “If any other diagnosis was written in the discharge diagnosis before one of our targets, the patient’s case was removed from the dataset.”
- ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution (2024, positive): “Interleaving LLM generations with evaluations yields powerful methods for prompt optimization (Zhou et al., 2022; Wang et al., 2023d; Guo et al., 2023), reinforcement learning (RL) reward design (Ma et al., 2023b), algorithmic (self-)improvement (Zelikman et al., 2023; Liu et al., 2023c;a), neural…”