Personalized response selection systems are generally grounded on persona. However, there exists a co-relation between persona and empathy, which is not explored well in these systems. Also, faithfulness to the conversation context plunges when a contradictory or an off-topic response is selected. This paper attempts to address these issues by proposing a suite of fusion strategies that capture the interaction between persona, emotion, and entailment information of the utterances. Ablation studies on the Persona-Chat dataset show that incorporating emotion and entailment improves the accuracy of response selection. We combine our fusion strategies and concept-flow encoding to train a BERT-based model which outperforms the previous methods by margins larger than 2.3 % on original personas and 1.9 % on revised personas in terms of hits@1 (top-1 accuracy), achieving a new state-of-the-art performance on the Persona-Chat dataset.
@article{arxiv.2208.09601,
title = {Using Multi-Encoder Fusion Strategies to Improve Personalized Response Selection},
author = {Souvik Das and Sougata Saha and Rohini K. Srihari},
journal= {arXiv preprint arXiv:2208.09601},
year = {2022}
}
Comments
Accepted by COLING 2022. arXiv admin note: text overlap with arXiv:2105.09050 by other authors