Learning from Text and Narrative Coherence
Su and Kyle discuss the importance of incorporating information from human crowdsource and the limitations of learning from text only. They explore the challenges in teaching models to tell coherent stories and the need for interpretability in downstream applications.In this clip
From this podcast
Data Skeptic
Interpretable One Shot Learning
Related Questions
Can language models understand stories as discussed in the episode Interpretable One Shot Learning and the clip Learning from Text and Narrative Coherence?
Can language models understand stories?
How does the language model work in the episode 61 - Neural Text Generation in Stories, with Elizabeth Clark and Yangfeng Ji and the clip Entity Representation in NLP?