[Submitted on 28 Apr 2026 (v1), last revised 24 Oct 2026 (this version, v3)]
Abstract:Large language models (LLMs) are increasingly used in emotionally sensitive human-AI applications, where reliable emotion detection is essential. However, their emotion recognition abilities remain uneven: models often perform well on some emotions while consistently struggling with others. Although recent work has explored emotion mechanisms in LLMs, little is known about why models are weaker on some emotions than others from a mechanistic interpretability perspective. In this work, we investigate emotion-specific biases through the causal mechanisms of emotion inference using sparse autoencoders (SAEs). We systematically identify causal sparse emotion features that drive emotion inference and analyze their sparse causal organization within and across emotions. We show that some emotions, such as surprise and fear, rely on highly concentrated feature sets, whereas disgust exhibits a more distributed sparse causal organization: its causal features are generally weaker, frequently co-activate with features for other emotions, and are often overshadowed by causal features for anger. These representational differences provide a mechanistic explanation for why LLMs struggle more with certain emotions. Finally, we conduct two intervention experiments: targeted steering of weaker causal features to mitigate emotion-specific failures, and global optimization of a steering vector over the identified causal features to improve overall emotion recognition performance.
Submission history
From: Bangzhao Shu [view email]
[v1] Tue, 28 Apr 2026 17:03:16 UTC (875 KB)
[v2] Thu, 25 Jun 2026 15:30:23 UTC (5,344 KB)
[v3] Sat, 24 Oct 2026 06:28:43 UTC (5,359 KB)
