AIBullisharXiv – CS AI · 14h ago6/10
🧠
A Language-Guided Bayesian Optimization for Efficient LoRA Hyperparameter Search
Researchers propose a Bayesian Optimization framework that uses pre-trained Large Language Models to efficiently search for optimal LoRA (Low-Rank Adaptation) hyperparameters by encoding domain knowledge as natural language prompts. The method discovers high-performing configurations in ~30 iterations versus 45,000 combinations, achieving 20% performance improvements while significantly reducing computational costs.