Evolution Fine-Tuning: Learning to Discover Across 371 Optimization Tasks A mid-training «practice phase» that teaches small open-source LLMs how to evolve solutions. Finch-8B is the 8B member of the Finch family — open-source LLMs evolution fine-tuned (EFT) to act as a stronger mutation operator inside evolutionary search. Uniquely in the family it is built on Qwen3-8B (the others use the Qwen3.5 series) and trained on the Finch Collection. It learns how to evolve a solution and is the strongest variant to pair with test-time RL. State-of-the-art discovery systems put an LLM inside an evolutionary search scaffold — but the discovery know-how lives in the scaffold, and every new task starts from zero. Evolution Fine-Tuning (EFT) moves that behavior into the model by turning evolutionary search trajectories into supervision. Finch-8B is a strong drop-in mutation operator that also synergizes with test-time reinforcement learning. — (Left) EFT acts as mid-training, boosting Finch’s discovery on the Erdős minimum-overlap problem under both test-time search and test-time learning. — (Right) On NP-hard competitive programming, Finch composes strategies learned across diverse domains,…
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: minnesotanlp
Теги: qwen3, evolution-fine-tuning, evolutionary-search, discovery, code-optimization, scientific-discovery, mutation-operator, mid-training
Лайков: 3 | Загрузок: 32
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
