Research
From Syntax to Emotion: A Mechanistic Analysis of Emotion Inference in LLMs
arXiv:2604.25866v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used in emotionally sensitive human-AI applications, yet little is known about how emotion recognition is
arXiv:2604.25866v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used in emotionally sensitive human-AI applications, yet little is known about how emotion recognition is internally represented. In this work, we investigate the internal mechanisms of emotion recognition in LLMs using sparse autoencoders (SAEs). By analyzing sparse feature activations across layers, we identify a consistent three-phase information flow, in which emotion-related features emerge only in the final phase. We further show that emotion representations comprise both shared features across emotions and emotion-specific features. Using phase-stratified causal tracing, we identify a small set of features that strongly influence emotion predictions, and show that both their number and causal impact vary across emotions; in particular, Disgust is more weakly and diffusely represented than other emotions. Finally, we propose an interpretable and data-efficient causal feature steering method that significantly improves emotion recognition performance across multiple models while largely preserving language modeling ability, and demonstrate that these improvements generalize across multiple emotion recognition datasets. Overall, our findings provide a systematic analysis of the internal mechanisms underlying emotion recognition in LLMs and introduce an efficient, interpretable, and controllable approach for improving model performance.
Related
- Contrastive Attribution in the Wild: An Interpretability Analysis of LLM Failures on Realistic Benchmarks
- From Fallback to Frontline: When Can LLMs be Superior Annotators of Human Perspectives?
- A closer look at how large language models trust humans: patterns and biases
- Preference Heads in Large Language Models: A Mechanistic Framework for Interpretable Personalization
- How Hypocritical Is Your LLM judge? Listener-Speaker Asymmetries in the Pragmatic Competence of Large Language Models
Source: arXiv cs.CL | 2026-04-29