3 papers
cs.CL2026
Why Are Some Emotions Harder for LLMs? Uncovering the Causal Mechanisms of Emotion Inference via Sparse Autoencoders
Bangzhao Shu, Arinjay Singh, Mai ElSherief
Large language models (LLMs) are increasingly used in emotionally sensitive human-AI applications, where reliable emotion detection is essential. However, their emotion recognition…
cs.SI2025
Large-Scale Analysis of Online Questions Related to Opioid Use Disorder on Reddit
Tanmay Laud, Akadia Kacha-Ochana, Steven A. Sumner +5
Opioid use disorder (OUD) is a leading health problem that affects individual well-being as well as general public health. Due to a variety of reasons, including the stigma faced b…
cs.SI2025
Exposure to Content Written by Large Language Models Can Reduce Stigma Around Opioid Use Disorder in Online Communities
Shravika Mittal, Darshi Shah, Shin Won Do +3
Widespread stigma, both in the offline and online spaces, acts as a barrier to harm reduction efforts in the context of opioid use disorder (OUD). This stigma is prominently direct…