Go Beyond Plain Fine-tuning: Improving Pretrained Models for Social Commonsense

Abstract: Pretrained LLMs have demonstrated outstanding performance in many NLP tasks recently. However, their social intelligence, which requires commonsense reasoning about the current situation and mental states of others, is still developing. Towards improving LLMs' social intelligence, we focus on the Social IQA dataset, a task requiring social and emotional commonsense reasoning. Building on top of the pretrained RoBERTa and GPT2 models, we propose several architecture variations and extensions, as well as leveraging external commonsense corpora, to optimize the model for Social IQA. Our proposed system achieves competitive results as those top-ranking models on the leaderboard. This work demonstrates the strengths of pretrained LLMs, and provides viable ways to improve their performance for a particular task.