← Home
Yejin Choi discusses her research on making small language models reason more effectively, challenging the assumption that scale is the only path to intelligence.
This summary was generated from show notes and public descriptions, not from a full transcript review. Details may contain inaccuracies.
Canon
•
Choi argues that the constraints of small models create a research environment that forces more creative algorithmic solutions than the brute-force scaling of large models.
•
Choi's research on small model reasoning challenges the dominant scaling hypothesis. Arguing that bigger is not always better in AI requires intellectual courage when the field is dominated by scale-focused labs.
Was this useful?