Emo Pillars is a synthetic 400K-utterance, 28-class emotion dataset generated by Mistral-7B from narrative plots; fine-tuned RoBERTa/BERT models achieve SOTA or competitive F1 on GoEmotions, ISEAR, IEMOCAP, and EmoContext.
What’s going on at the Tyrell Corporation? Why are these replicants here and what do they want from Tyrell?
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Emo Pillars: Knowledge Distillation to Support Fine-Grained Context-Aware and Context-Less Emotion Classification
Emo Pillars is a synthetic 400K-utterance, 28-class emotion dataset generated by Mistral-7B from narrative plots; fine-tuned RoBERTa/BERT models achieve SOTA or competitive F1 on GoEmotions, ISEAR, IEMOCAP, and EmoContext.