Large Language Models (LLMs) have showcased remarkable capabilities in various Natural Language Processing tasks. For automatic open-domain dialogue evaluation in particular, LLMs have been seamlessly integrated into evaluation frameworks, and together with human evaluation, compose the backbone of most evaluations. However, existing evaluation benchmarks often rely on outdated datasets and evaluate aspects like Fluency and Relevance, which fail to adequately capture the capabilities and limitations of state-of-the-art chatbot models. This paper critically examines current evaluation benchmarks, highlighting that the use of older response generators and quality aspects fail to accurately reflect modern chatbot capabilities. A small annotation experiment on a recent LLM-generated dataset (SODA) reveals that LLM evaluators such as GPT-4 struggle to detect actual deficiencies in dialogues generated by current LLM chatbots.

Although human evaluation remains the gold standard for open-domain dialogue evaluation, the growing popularity of automated evaluation using Large Language Models (LLMs) has also extended to dialogue. However, most frameworks leverage benchmarks that assess older chatbots on aspects such as fluency and relevance, which are not reflective of the challenges associated with contemporary models. In fact, a qualitative analysis on Soda. (Kim et al., 2023), a GPT-3.5 generated dialogue dataset, suggests that current chatbots may exhibit several recurring issues related to coherence and commonsense knowledge, but generally produce highly fluent and relevant responses.Noting the aforementioned limitations, this paper introduces Soda-Eval, an annotated dataset based on Soda that covers over 120K turn-level assessments across 10K dialogues, where the annotations were generated by GPT-4. Using Soda-Eval as a benchmark, we then study the performance of several open-access instruction-tuned LLMs, finding that dialogue evaluation remains challenging. Fine-tuning these models improves performance over few-shot inferences, both in terms of correlation and explanation.

Despite being heralded as the new standard for dialogue evaluation, the closed-source nature of GPT-4 poses challenges for the community. Motivated by the need for lightweight, open source, and multilingual dialogue evaluators, this paper introduces GenResCoh (Generated Responses targeting Coherence). GenResCoh is a novel LLM generated dataset comprising over 130k negative and positive responses and accompanying explanations seeded from XDailyDialog and XPersona covering English, French, German, Italian, and Chinese. Leveraging GenResCoh, we propose ECoh (Evaluation of Coherence), a family of evaluators trained to assess response coherence across multiple languages. Experimental results demonstrate that ECoh achieves multilingual detection capabilities superior to the teacher model (GPT-3.5-Turbo) on GenResCoh, despite being based on a much smaller architecture. Furthermore, the explanations provided by ECoh closely align in terms of quality with those generated by the teacher model.

The main limiting factor in the development of robust multilingual open-domain dialogue evaluation metrics is the lack of multilingual data and the limited availability of open-sourced multilingual dialogue systems. In this work, we propose a workaround for this lack of data by leveraging a strong multilingual pretrained encoder-based Language Model and augmenting existing English dialogue data using Machine Translation. We empirically show that the naive approach of finetuning a pretrained multilingual encoder model with translated data is insufficient to outperform the strong baseline of finetuning a multilingual model with only source data. Instead, the best approach consists in the careful curation of translated data using MT Quality Estimation metrics, excluding low quality translations that hinder its performance.

Despite significant research effort in the development of automatic dialogue evaluation metrics, little thought is given to evaluating dialogues other than in English. At the same time, ensuring metrics are invariant to semantically similar responses is also an overlooked topic. In order to achieve the desired properties of robustness and multilinguality for dialogue evaluation metrics, we propose a novel framework that takes advantage of the strengths of current evaluation models with the newly-established paradigm of prompting Large Language Models (LLMs). Empirical results show our framework achieves state of the art results in terms of mean Spearman correlation scores across several benchmarks and ranks first place on both the Robust and Multilingual tasks of the DSTC11 Track 4 “Automatic Evaluation Metrics for Open-Domain Dialogue Systems”, proving the evaluation capabilities of prompted LLMs.

Task-oriented conversational datasets often lack topic variability and linguistic diversity. However, with the advent of Large Language Models (LLMs) pretrained on extensive, multilingual and diverse text data, these limitations seem overcome. Nevertheless, their generalisability to different languages and domains in dialogue applications remains uncertain without benchmarking datasets. This paper presents a holistic annotation approach for emotion and conversational quality in the context of bilingual customer support conversations. By performing annotations that take into consideration the complete instances that compose a conversation, one can form a broader perspective of the dialogue as a whole. Furthermore, it provides a unique and valuable resource for the development of text classification models. To this end, we present benchmarks for Emotion Recognition and Dialogue Quality Estimation and show that further research is needed to leverage these models in a production setting.

Despite considerable advances in open-domain neural dialogue systems, their evaluation remains a bottleneck. Several automated metrics have been proposed to evaluate these systems, however, they mostly focus on a single notion of quality, or, when they do combine several sub-metrics, they are computationally expensive. This paper attempts to solve the latter: QualityAdapt leverages the Adapter framework for the task of Dialogue Quality Estimation. Using well defined semi-supervised tasks, we train adapters for different subqualities and score generated responses with AdapterFusion. This compositionality provides an easy to adapt metric to the task at hand that incorporates multiple subqualities. It also reduces computational costs as individual predictions of all subqualities are obtained in a single forward pass. This approach achieves comparable results to state-of-the-art metrics on several datasets, whilst keeping the previously mentioned advantages.

Crowdsourcing the collection of speech provides a scalable setting to access a customisable demographic according to each dataset’s needs. The correctness of speaker metadata is especially relevant for speaker-centred collections - ones that require the collection of a fixed amount of data per speaker. This paper identifies two different types of misalignment present in these collections: Multiple Accounts misalignment (different contributors map to the same speaker), and Multiple Speakers misalignment (multiple speakers map to the same contributor). Based on state-of-the-art approaches to Speaker Verification, this paper proposes an unsupervised method for measuring speaker metadata plausibility of a collection, i.e., evaluating the match (or lack thereof) between contributors and speakers. The solution presented is composed of an embedding extractor and a clustering module. Results indicate high precision in automatically classifying contributor alignment (>0.94).