The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) cover image

BloombergGPT - an LLM for Finance with David Rosenberg - #639

The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

00:00

Evaluating Financial Language Models

This chapter focuses on the evaluation process of trained financial language models, exploring validation methods and performance benchmarks. It highlights the importance of data selection, tokenizer strategies, and the balance between training from scratch and fine-tuning existing models. The discussion also includes challenges in generative tasks, the significance of instruction tuning, and the implications of training techniques on model efficiency.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app