Dexa
/
NLP Highlights
Learn more
Follow
NLP Model Analysis
Matt and Afra discuss the importance of using informed baselines in NLP tasks to avoid misleading correlations. Tal highlights a paper exploring how to ensure models are utilizing information in their internal representations, not just encoding it.
Add to Radar
Share
In this clip
Matt Gardner
Afra Alishahi
Tal Linzen
From this podcast
NLP Highlights
81 - BlackboxNLP, with Afra Alishahi and Tal Linzen
Related Questions
How does the language model diagnostic classifier work in the episode 81 - BlackboxNLP, with Afra Alishahi and Tal Linzen and the clip Language Model Diagnostic Classifier?
