Mergenetic: a Simple Evolutionary Model Merging Library
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912547837313024 |
|---|---|
| author | Minut, Adrian Robert Mencattini, Tommaso Santilli, Andrea Crisostomi, Donato Rodolà, Emanuele |
| author_facet | Minut, Adrian Robert Mencattini, Tommaso Santilli, Andrea Crisostomi, Donato Rodolà, Emanuele |
| contents | Model merging allows combining the capabilities of existing models into a new one - post hoc, without additional training. This has made it increasingly popular thanks to its low cost and the availability of libraries that support merging on consumer GPUs. Recent work shows that pairing merging with evolutionary algorithms can boost performance, but no framework currently supports flexible experimentation with such strategies in language models. We introduce Mergenetic, an open-source library for evolutionary model merging. Mergenetic enables easy composition of merging methods and evolutionary algorithms while incorporating lightweight fitness estimators to reduce evaluation costs. We describe its design and demonstrate that Mergenetic produces competitive results across tasks and languages using modest hardware. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_11427 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Mergenetic: a Simple Evolutionary Model Merging Library Minut, Adrian Robert Mencattini, Tommaso Santilli, Andrea Crisostomi, Donato Rodolà, Emanuele Machine Learning Artificial Intelligence Neural and Evolutionary Computing Model merging allows combining the capabilities of existing models into a new one - post hoc, without additional training. This has made it increasingly popular thanks to its low cost and the availability of libraries that support merging on consumer GPUs. Recent work shows that pairing merging with evolutionary algorithms can boost performance, but no framework currently supports flexible experimentation with such strategies in language models. We introduce Mergenetic, an open-source library for evolutionary model merging. Mergenetic enables easy composition of merging methods and evolutionary algorithms while incorporating lightweight fitness estimators to reduce evaluation costs. We describe its design and demonstrate that Mergenetic produces competitive results across tasks and languages using modest hardware. |
| title | Mergenetic: a Simple Evolutionary Model Merging Library |
| topic | Machine Learning Artificial Intelligence Neural and Evolutionary Computing |
| url | https://arxiv.org/abs/2505.11427 |