Towards high-quality (maybe synthetic) datasets (Practical AI #290)
As Argilla puts it: "Data quality is what makes or breaks AI." However, what exactly does this mean and how can AI team probably collaborate with domain experts towards improved data quality? David Berenstein & Ben Burtenshaw, who are building Argilla & Distilabel at Hugging Face, join us to dig into these topics along with synthetic data generation & AI-generated labeling / feedback.
Embed this episode
Chapters
- 0:00 Welcome to Practical AI
- 0:44 Sponsor: Fly
- 3:56 What does data collaboration mean?
- 7:18 Understanding your data
- 9:58 How to start curating data
- 13:12 Practical steps to scale
- 16:52 Sponsor: WorkOS
- 20:23 Traditional & new usecases
- 24:51 Virtues of smaller models
- 27:04 What Argilla looks like
- 30:55 User backgrounds
- 34:21 The non-technical POV
- 38:23 Sponsor: Eight Sleep
- 41:09 AI feedback
- 44:50 Hallucination issues
- 46:10 What is Distilabel
- 50:08 Usage & adoption
- 52:55 Where things are going
- 55:34 This is muy bueno
- 56:15 Outro
No transcript yet.
Frequently asked questions
How long is this episode of Changelog Master Feed?
This episode is 57 minutes long.
When was this Changelog Master Feed episode published?
This episode was published on October 9, 2024.
Is there a transcript available for this episode?
Yes, a full transcript is available for this episode. You can read the complete transcript on the episode page.
Can I download this Changelog Master Feed episode?
Yes. Use the download control on the episode player to save the publisher-provided media file.