Redundancy, Isotropy, and Intrinsic Dimensionality of Prompt-based Text Embeddings

By Hayato Tsukagoshi, Ryohei Sasano. 2025

Subfields: Natural Language Processing, Representation & Self-Supervised Learning

Citations in the corpus

• When Is 0.1% Enough? Analyzing the Combined Effects of Dimensionality Reduction and Quantization on Text Embedding Compression (2026, unknown): “We evaluate four pretrained text embedding models, following the model selection strategy of prior work on the dimensional redundancy of prompt-based text embeddings (Tsukagoshi and Sasano, 2025).”