Completed
00:00:01 - Intro & Webinar Overview
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Improving the Accuracy of Domain Specific Tasks with LLM Distillation
Automatically move to the next video in the Classroom when playback concludes
- 1 00:00:01 - Intro & Webinar Overview
- 2 00:03:19 - The Cost Challenge of LLMs
- 3 00:06:17 - Complexity in Real-World AI Tasks
- 4 00:09:34 - Evaluating Domain-Specific Accuracy
- 5 00:12:41 - Ranking and Response Generation
- 6 00:15:58 - Using DistilBERT and Model Combinations
- 7 00:19:25 - Tools for Efficient LLM Distillation
- 8 00:22:45 - Architecture of Distilled Models
- 9 00:26:11 - Presentation Format and Flow
- 10 00:29:34 - Use of Labeled and Weak Data
- 11 00:32:57 - Tradeoffs in Deploying Large Models
- 12 00:36:15 - Case Study: Enterprise Deployment
- 13 00:39:28 - Generalization vs Specialization
- 14 00:42:52 - Model Size and Performance Comparison
- 15 00:46:20 - Final Thoughts & Q&A
