Embedding Models Measure in Peculiar Ways

2026-09-17

AI Summary

研究探讨了嵌入模型在表达物理量度（如质量、距离、时间、体积）时的局限性，发现这些模型受表面字符串相似性影响较大，重新校准相似度也无法显著改善。

Abstract

Embedding spaces define notions of semantic similarity and distance. We study whether those embeddings reflect physical measurements of mass, distance, time and volume, which admit a unique, objective notion of semantic equivalence and distance. We find that physical measurement is only weakly modeled in the embedding space, and that instead quite peculiar measurement patterns can be observed. Further analysis indicates that embedding representations of physical measurements are strongly influenced by superficial string similarity, and recalibration of similarity does not substantially improve the alignment.

Problem

• embedding models
• physical measurements
• semantic similarity
• distance

Innovation

• embedding spaces
• physical measurements
• semantic similarity
• string similarity