Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning

2 points| ijk | 4 months ago |arxiv.org