正常,给各种LLM打分标准是啥(有人做了但不尽人意或者迷糊),那么打造LLM训练标准又是如何?这个更模糊🤔
现在每天都一堆LLM 数据更是堆积如山,莫非让其左右互搏?那有如何进行,🤔
严重超纲
可以水无数论文🥸
https://www.rainforestqa.com/blog/building-reliable-systems-out-of-unreliable-agents
Rainforest QA Blog | Software Testing Guides
Building reliable systems out of unreliable agents | Rainforest QA
This is the process our engineering team uses to create reliable AI systems out of unreliable AI agents.
via MJJ出征 - Telegram Channel