paper-with-me

홈 › Papers

Language Model Transformers as Evaluators for Open-domain Dialogues

2020-12-01 · COLING 2020 8 · Rostislav Nedelchev, Jens Lehmann, Ricardo Usbeck

Computer-based systems for communication with humans are a cornerstone of AI research since the 1950s. So far, the most effective way to assess the quality of the dialogues produced by these systems is to use resource-intensive manual labor instead of automated means. In this work, we investigate whether language models (LM) based on transformer neural networks can indicate the quality of a conversation. In a general sense, language models are methods that learn to predict one or more words based on an already given context. Due to their unsupervised nature, they are candidates for efficient, automatic indication of dialogue quality. We demonstrate that human evaluators have a positive correlation between the output of the language models and scores. We also provide some insights into their behavior and inner-working in a conversational context.

📄 PDF Abstract BibTeX

Code (1)

smartdataanalytics/transformers_dialogue_evaluators 공식 구현 pytorch

Tasks

Language ModelingLanguage Modelling

Similar Papers 제목 키워드 기반

MEDAL: A Framework for Benchmarking LLMs as Multilingual Open-Domain Chatbots and Dialogue Evaluators

2025-05-28 · John Mendonça, Alon Lavie, Isabel Trancoso

As the capabilities of chatbots and their underlying LLMs continue to dramatically improve, evaluating their performance has increasingly become a major blocker to their further development. A major challenge is the avai…

BenchmarkingChatbotDialogue Evaluation

On the Benchmarking of LLMs for Open-Domain Dialogue Evaluation

2024-07-04 · John Mendonça, Alon Lavie, Isabel Trancoso

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 eva…

BenchmarkingChatbotDialogue Evaluation

ECoh: Turn-level Coherence Evaluation for Multilingual Dialogues

2024-07-16 · John Mendonça, Isabel Trancoso, Alon Lavie

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 eval…

Coherence EvaluationDialogue Evaluation

Proxy Indicators for the Quality of Open-domain Dialogues

2021-11-01 · EMNLP 2021 11 · Rostislav Nedelchev, Jens Lehmann, Ricardo Usbeck

The automatic evaluation of open-domain dialogues remains a largely unsolved challenge. Despite the abundance of work done in the field, human judges have to evaluate dialogues’ quality. As a consequence, performing such…

Dialogue Evaluation

Are Pre-trained Language Models Knowledgeable to Ground Open Domain Dialogues?

2020-11-19 · Yufan Zhao, Wei Wu, Can Xu

We study knowledge-grounded dialogue generation with pre-trained language models. Instead of pursuing new state-of-the-art on benchmarks, we try to understand if the knowledge stored in parameters of the pre-trained mode…

Dialogue Generation