Statistical Depth for Ranking and Characterizing Transformer-Based Text Embeddings

Parker Seegmiller, Sarah Masud Preum
December 2023
PDF Cite Code
Statistical Depth

Abstract

A statistical tool for measuring and comparing distributions of transformer-based text embeddings.

Type
Conference paper
Publication
The 2023 Conference on Empirical Methods in Natural Language Processing
Parker Seegmiller
Parker Seegmiller
Innovation Fellow, PhD Candidate @ Dartmouth CS | Applied Science Intern @ Amazon

nlp @ dartmouth college, llms @ amazon