3 ms·Fine-tuning vs. in-context learning: LLM customization for real-world tasks4 points by Bluestein 1y ago