Using the Same Training Set in the Large Language Models

The same training set should be used for all. Because it's providing kind of the broadest coverage. So like, for example, when you're looking at the IVR space, someone still could say credit card, right? And so you still want to be able to handle some of these types of short utterances that you may get in like in a chat in the IVR. It's really your testing and your, your batch tests, utterances and things like that that tend to vary by channel. But your training typically is consistent and should be consistent across all.

Transcript

Play full episode

Remember Everything You Learn from Podcasts

Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.