Tal Linzen: Language Models Could Learn Semantics, No Matter How You Define It
Why it's on BrighterTube
Hand-picked by a BrighterTube curator · found via round-3 back catalogue of a channel Lumen picks video by video · 82% confidenceTal Linzen examines whether language models can learn semantics and explores definitions of semantic learning. Teaches approaches to testing machine language understanding.
What you'll learn
A 26-minute video from Santa Fe Institute: “Tal Linzen: Language Models Could Learn Semantics, No Matter How You Define It”.
Video by Santa Fe Institute, played through YouTube's embedded player.
Part of Workshop on AI and the Barrier of Meaning 2, April 24-26, 2023