# Evaluating Coherence in Dialogue Systems using Entailment

## TL;DR

This paper proposes new automatic, interpretable metrics based on entailment and sentence representations to evaluate dialogue coherence, aiming to replace costly human judgments and improve scalability and consistency.

## Contribution

It introduces novel entailment-based metrics for assessing dialogue coherence, providing a scalable and unbiased alternative to human evaluation.

## Key findings

- Metrics correlate well with human judgments
- Enables large-scale evaluation of dialogue systems
- Reduces evaluation bias and cost

## Abstract

Evaluating open-domain dialogue systems is difficult due to the diversity of possible correct answers. Automatic metrics such as BLEU correlate weakly with human annotations, resulting in a significant bias across different models and datasets. Some researchers resort to human judgment experimentation for assessing response quality, which is expensive, time consuming, and not scalable. Moreover, judges tend to evaluate a small number of dialogues, meaning that minor differences in evaluation configuration may lead to dissimilar results. In this paper, we present interpretable metrics for evaluating topic coherence by making use of distributed sentence representations. Furthermore, we introduce calculable approximations of human judgment based on conversational coherence by adopting state-of-the-art entailment techniques. Results show that our metrics can be used as a surrogate for human judgment, making it easy to evaluate dialogue systems on large-scale datasets and allowing an unbiased estimate for the quality of the responses.