Lyria: A General LLM - Driven Genetic Algorithm Framework for Problem Solving
Weizhi Tang, Kwabena Nuamah, Vaishak Belle · 2025
Read Lyria: A General LLM - Driven Genetic Algorithm Framework for Problem Solving by Weizhi Tang, Kwabena Nuamah, Vaishak Belle (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-04034.txt.
Description
Weizhi Tang, Kwabena Nuamah, and Vaishak Belle present Lyria, a framework that pairs language models with genetic search, in this 2025 paper. Language models get stuck in local optima and skip parts of the solution space. Genetic algorithms explore broadly but lack language understanding. Lyria joins them with symbolic tools in seven components.
Experiments ran four language models on three problem types. Ablations tested which parts of the system matter. The combined search beat plain model reasoning in their trials. A follow on method, LAFT, lets a weaker model imitate the stronger system reasoning traces. It beat nine baselines in their comparisons. Evolution style search may patch a known blind spot of language models.
Catalog record
| CREATOR | Weizhi Tang, Kwabena Nuamah, Vaishak Belle |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 98 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-04034