StressRoBERTa: Cross-Condition Transfer Learning for Stress Detection
Published:Dec 29, 2025 19:16
•1 min read
•ArXiv
Analysis
This paper is significant because it addresses the challenge of detecting chronic stress on social media, a growing public health concern. It leverages transfer learning from related mental health conditions (depression, anxiety, PTSD) to improve stress detection accuracy. The results demonstrate the effectiveness of this approach, outperforming existing methods and highlighting the value of focused cross-condition training.
Key Takeaways
- •Proposes StressRoBERTa, a cross-condition transfer learning model for stress detection.
- •Utilizes RoBERTa and pre-trains on data from depression, anxiety, and PTSD.
- •Achieves state-of-the-art results on the SMM4H 2022 Task 8 dataset.
- •Demonstrates the effectiveness of transfer learning from related mental health conditions for stress detection.
Reference
“StressRoBERTa achieves 82% F1-score, outperforming the best shared task system (79% F1) by 3 percentage points.”