Towards a Multidimensional Evaluation Framework for Empathetic Conversational Systems
This work addresses the need for better evaluation metrics in empathetic conversational systems, which is an incremental improvement for researchers and developers in AI and human-computer interaction.
The paper tackled the problem of inadequate evaluation methods for empathetic conversational systems by proposing a multidimensional framework with three new methods to measure empathy at structural, behavioral, and overall levels, and demonstrated its usefulness through experiments with state-of-the-art models and LLMs.
Empathetic Conversational Systems (ECS) are built to respond empathetically to the user's emotions and sentiments, regardless of the application domain. Current ECS studies evaluation approaches are restricted to offline evaluation experiments primarily for gold standard comparison & benchmarking, and user evaluation studies for collecting human ratings on specific constructs. These methods are inadequate in measuring the actual quality of empathy in conversations. In this paper, we propose a multidimensional empathy evaluation framework with three new methods for measuring empathy at (i) structural level using three empathy-related dimensions, (ii) behavioral level using empathy behavioral types, and (iii) overall level using an empathy lexicon, thereby fortifying the evaluation process. Experiments were conducted with the state-of-the-art ECS models and large language models (LLMs) to show the framework's usefulness.