For humans and computers to communicate via natural language text, it is necessary that the understanding (interpretation) of the given texts be shared. In this research, we tackle the issues of dialogue system and language grounding.
Evaluating Quality of a Dialogue Response
Evaluating natural language is a challenging task. Without a proper assessment of a text’s quality, it is difficult to determine which models can produce a better generation. In this research, we focus on techniques to enhance the evaluation of NLG.
Many automatic evaluation metrics have been proposed to score the overall quality of a response in open-domain dialogue. Generally, the overall quality is comprised of various aspects, such as relevancy, specificity, and empathy, and the importance of each aspect differs according to the task. For instance, specificity is mandatory in a food-ordering dialogue task, whereas fluency is preferred in a language-teaching dialogue system. However, existing metrics are not designed to cope with such flexibility. For example, BLEU score fundamentally relies only on word overlapping, whereas BERTScore relies on semantic similarity between reference and candidate response. Thus, they are not guaranteed to capture the required aspects, i.e., specificity. To design a metric that is flexible to a task, we first propose making these qualities manageable by grouping them into three groups: understandability, sensibleness, and likability, where likability is a combination of qualities that are essential for a task. We also propose a simple method to composite metrics of each aspect to obtain a single metric called USL-H, which stands for Understandability, Sensibleness, and Likability in Hierarchy. We demonstrated that USL-H score achieves good correlations with human judgment and maintains its configurability towards different aspects and metrics. [ Vitou Phy et al.: COLING-2020 , https://github.com/vitouphy/usl_dialogue_metric ]
A Natural Language Corpus of Common Grounding under Continuous and Partially-Observable Context
Common grounding is the process of creating, repairing and updating mutual understandings, which is a critical aspect of sophisticated human communication. However, traditional dialogue systems have limited capability of establishing common ground, and we also lack task formulations which introduce natural difficulty in terms of common grounding while enabling easy evaluation and analysis of complex models. In this work, we propose a minimal dialogue task which requires advanced skills of common grounding under continuous and partially-observable context. Based on this task formulation, we collected a largescale dataset of 6,760 dialogues which fulfills essential requirements of natural language corpora. Our analysis of the dataset revealed important phenomena related to common grounding that need to be considered. Finally, we evaluate and analyze baseline neural models on a simple subtask that requires recognition of the created common ground. We show that simple baseline models perform decently but leave room for further improvement. Overall, we show that our proposed task will be a fundamental testbed where we can train, evaluate, and analyze dialogue system’s ability for sophisticated common grounding. (Udagawa et al.: AAAI-2019, Udagawa et al.: AAAI-2020, https://github.com/Alab-NII/onecommon)