PMIScore: An Unsupervised Approach to Quantify Dialogue Engagement
This provides a method to benchmark large language models, enhance human-computer interactions, or improve personal communication skills, but it is incremental as it adapts existing PMI concepts to dialogue.
The paper tackled the problem of quantifying dialogue engagement, which is subjective and lacks a gold standard, by proposing PMIScore, an unsupervised approach using pointwise mutual information (PMI), and validated it on synthetic and real-world datasets to demonstrate effectiveness and reasonableness.
High dialogue engagement is a crucial indicator of an effective conversation. A reliable measure of engagement could help benchmark large language models, enhance the effectiveness of human-computer interactions, or improve personal communication skills. However, quantifying engagement is challenging, since it is subjective and lacks a "gold standard". This paper proposes PMIScore, an efficient unsupervised approach to quantify dialogue engagement. It uses pointwise mutual information (PMI), which is the probability of generating a response conditioning on the conversation history. Thus, PMIScore offers a clear interpretation of engagement. As directly computing PMI is intractable due to the complexity of dialogues, PMIScore learned it through a dual form of divergence. The algorithm includes generating positive and negative dialogue pairs, extracting embeddings by large language models (LLMs), and training a small neural network using a mutual information loss function. We validated PMIScore on both synthetic and real-world datasets. Our results demonstrate the effectiveness of PMIScore in PMI estimation and the reasonableness of the PMI metric itself.