Across different architectures, all AI models represent emotions using the exact same mathematical shape.
April 14, 2026
Original Paper
Shared Emotion Geometry Across Small Language Models: A Cross-Architecture Study of Representation, Behavior, and Methodological Confounds
arXiv · 2604.11050
The Takeaway
Five distinct LLM architectures share a nearly identical internal geometry for 21 emotions. This suggests that 'emotion' is a universal mathematical structure that emerges naturally whenever a model processes human language.
From the abstract
We extract 21-emotion vector sets from twelve small language models (six architectures x base/instruct, 1B-8B parameters) under a unified comprehension-mode pipeline at fp16 precision, and compare the resulting geometries via representational similarity analysis on raw cosine RDMs. The five mature architectures (Qwen 2.5 1.5B, SmolLM2 1.7B, Llama 3.2 3B, Mistral 7B v0.3, Llama 3.1 8B) share nearly identical 21-emotion geometry, with pairwise RDM Spearman correlations of 0.74-0.92. This universal