PVG at WASSA 2021: A Multi-Input, Multi-Task, Transformer-Based Architecture for Empathy and Distress Prediction
This work addresses improving emotion prediction for human-machine interaction, but it is incremental as it builds on existing transformer-based methods with multi-task learning.
The paper tackled predicting empathy and distress scores from textual and ancillary data, achieving first place in average correlation (0.545) and distress correlation (0.574), and second in empathy correlation (0.517) in the WASSA 2021 shared task.
Active research pertaining to the affective phenomenon of empathy and distress is invaluable for improving human-machine interaction. Predicting intensities of such complex emotions from textual data is difficult, as these constructs are deeply rooted in the psychological theory. Consequently, for better prediction, it becomes imperative to take into account ancillary factors such as the psychological test scores, demographic features, underlying latent primitive emotions, along with the text's undertone and its psychological complexity. This paper proffers team PVG's solution to the WASSA 2021 Shared Task on Predicting Empathy and Emotion in Reaction to News Stories. Leveraging the textual data, demographic features, psychological test score, and the intrinsic interdependencies of primitive emotions and empathy, we propose a multi-input, multi-task framework for the task of empathy score prediction. Here, the empathy score prediction is considered the primary task, while emotion and empathy classification are considered secondary auxiliary tasks. For the distress score prediction task, the system is further boosted by the addition of lexical features. Our submission ranked 1$^{st}$ based on the average correlation (0.545) as well as the distress correlation (0.574), and 2$^{nd}$ for the empathy Pearson correlation (0.517).