AI RESEARCH
Valence-Arousal Subspace in LLMs: Circular Emotion Geometry and Multi-Behavioral Control
arXiv CS.AI
•
ArXi:2604.03147v1 Announce Type: cross We present a method to identify a valence-arousal (VA) subspace within large language model representations. From 211k emotion-labeled texts, we derive emotion steering vectors, then learn VA axes as linear combinations of their top PCA components via ridge regression on the model's self-reported valence-arousal scores. The resulting VA subspace exhibits circular geometry consistent with established models of human emotion perception. Projections along our recovered VA subspace correlate with human-crowdsourced VA ratings across 44k lexical items.