On the Transformation of Latent Space in Fine-Tuned NLP Models

Understand

We study the evolution of latent space in fine-tuned NLP models.

• Different from the commonly used probing-framework, we opt for an unsupervised method to analyze representations.
• More specifically, we discover latent concepts in the representational space using hierarchical clustering.
• We then use an alignment function to gauge the similarity between the latent space of a pre-trained model and its fine-tuned version.