Why Models Are AI’s Next Training Dataset with Damian Borth - #772
2026-07-27 · 47 min · episode 772 · 13 entities
Asserted relationships
-
→ appeared on The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence podcast0.68
evidence rules-v4
Why Models Are AI’s Next Training Dataset with Damian Borth - #772
-
0.55
evidence rules-v4
Feed author/publisher: Sam Charrington
-
0.40
evidence rules-v4
Feed category: Science
-
0.40
evidence rules-v4
Feed category: Technology
-
0.40
evidence rules-v4
Feed category: News
-
0.40
evidence rules-v4
Feed category: Tech News
-
0.40
evidence rules-v4
Feed author/publisher: TWIML
Entities found in this episode
concepts 5
-
0.50
evidence rules-v4
Feed category: Science
-
0.50
evidence rules-v4
Feed category: Tech News
-
0.40
evidence rules-v4
Feed category: Science
-
0.40
evidence rules-v4
Feed category: News
-
0.40
evidence rules-v4
Feed category: Tech News
companys 4
-
0.70
evidence rules-v4
Feed author/publisher: TWIML
-
0.50
evidence rules-v4
Feed category: Technology
-
0.40
evidence rules-v4
Feed category: Technology
-
0.40
evidence rules-v4
Feed author/publisher: TWIML
persons 3
-
0.72
evidence rules-v4
Why Models Are AI’s Next Training Dataset with Damian Borth - #772
-
0.70
evidence rules-v4
Feed author/publisher: Sam Charrington
-
0.55
evidence rules-v4
Feed author/publisher: Sam Charrington
podcasts 1
-
appeared on The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence podcast0.68
evidence rules-v4
Why Models Are AI’s Next Training Dataset with Damian Borth - #772
Episode description as stored
For more than a decade, AI has advanced by training ever-larger models on ever-larger datasets. But as high-quality training data becomes harder to find and pretraining grows increasingly expensive, researchers are looking for new ways to keep foundation models improving.
In this episode, Damian Borth, professor of AI and machine learning at the University of St. Gallen, argues we’ve been overlooking an important source of knowledge: the models we’ve already trained. His group’s work on weight space learning treats trained neural networks themselves as data, learning from the distilled results of millions of GPU hours of optimization rather than starting from raw data each time.
We explore what it means to build foundation models of neural networks, how knowledge can be transferred across architectures and domains, why this approach could dramatically reduce the cost of developing specialized models, and whether future AI systems may be trained on collections of existing models instead of ever-growing datasets.
🗒️ Full show notes: https://twimlai.com/go/772.