Revisiting the Role of Similarity and Dissimilarity in Best Counter Argument Retrieval
This work addresses the problem of efficient and effective counter-argument retrieval for natural language processing applications, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the task of retrieving the best counter-argument for a given input argument by developing a model that scores based on similarity and dissimilarity metrics, achieving an accuracy@1 of 49.04% and significantly outperforming baselines.
This paper studies the task of best counter-argument retrieval given an input argument. Following the definition that the best counter-argument addresses the same aspects as the input argument while having the opposite stance, we aim to develop an efficient and effective model for scoring counter-arguments based on similarity and dissimilarity metrics. We first conduct an experimental study on the effectiveness of available scoring methods, including traditional Learning-To-Rank (LTR) and recent neural scoring models. We then propose Bipolar-encoder, a novel BERT-based model to learn an optimal representation for simultaneous similarity and dissimilarity. Experimental results show that our proposed method can achieve the accuracy@1 of 49.04\%, which significantly outperforms other baselines by a large margin. When combined with an appropriate caching technique, Bipolar-encoder is comparably efficient at prediction time.