Large Language Models (LLMs) are on the headlines in many different fields. How to validate whether a LLM is good for the tasks you want to perform against? It is a daunting task as LLMs are unpredictable with hallucination. Every time you might get slightly different answers for the same prompt. In this session, Emily will share her experience on validating a Java LLM for test generation, code generation, Java migration, etc. In particular, the code generation with comment to code is daunting as you might get vastly different variations for doing the similar thing. Come to this session to learn how the metrics was applied.Level:Beginner
