Is chatgpt a good nlg evaluator? a preliminary study

J Wang, Y Liang, F Meng, Z Sun, H Shi, Z Li… - arXiv preprint arXiv …, 2023 - arxiv.org
Recently, the emergence of ChatGPT has attracted wide attention from the computational
linguistics community. Many prior studies have shown that ChatGPT achieves remarkable …

Bartscore: Evaluating generated text as text generation

W Yuan, G Neubig, P Liu - Advances in Neural Information …, 2021 - proceedings.neurips.cc
A wide variety of NLP applications, such as machine translation, summarization, and dialog,
involve text generation. One major challenge for these applications is how to evaluate …

TRUE: Re-evaluating factual consistency evaluation

O Honovich, R Aharoni, J Herzig, H Taitelbaum… - arXiv preprint arXiv …, 2022 - arxiv.org
Grounded text generation systems often generate text that contains factual inconsistencies,
hindering their real-world applicability. Automatic factual consistency evaluation may help …

Automatic machine translation evaluation in many languages via zero-shot paraphrasing

B Thompson, M Post - arXiv preprint arXiv:2004.14564, 2020 - arxiv.org
We frame the task of machine translation evaluation as one of scoring machine translation
output with a sequence-to-sequence paraphraser, conditioned on a human reference. We …

Natural language watermarking via paraphraser-based lexical substitution

J Qiang, S Zhu, Y Li, Y Zhu, Y Yuan, X Wu - Artificial Intelligence, 2023 - Elsevier
Although powerful pretrained language models generate high-quality output text, they bring
new concerns about the potential misuse of such models for malicious purposes. Natural …

Quality controlled paraphrase generation

E Bandel, R Aharonov, M Shmueli-Scheuer… - arXiv preprint arXiv …, 2022 - arxiv.org
Paraphrase generation has been widely used in various downstream tasks. Most tasks
benefit mainly from high quality paraphrases, namely those that are semantically similar to …

Data augmentation with paraphrase generation and entity extraction for multimodal dialogue system

E Okur, S Sahay, L Nachman - arXiv preprint arXiv:2205.04006, 2022 - arxiv.org
Contextually aware intelligent agents are often required to understand the users and their
surroundings in real-time. Our goal is to build Artificial Intelligence (AI) systems that can …

Paraphrase generation as zero-shot multilingual translation: Disentangling semantic similarity from lexical and syntactic diversity

B Thompson, M Post - arXiv preprint arXiv:2008.04935, 2020 - arxiv.org
Recent work has shown that a multilingual neural machine translation (NMT) model can be
used to judge how well a sentence paraphrases another sentence in the same language …

AutoQA: From databases to QA semantic parsers with only synthetic training data

S Xu, SJ Semnani, G Campagna, MS Lam - arXiv preprint arXiv …, 2020 - arxiv.org
We propose AutoQA, a methodology and toolkit to generate semantic parsers that answer
questions on databases, with no manual effort. Given a database schema and its data …

An investigation of evaluation methods in automatic medical note generation

AB Abacha, W Yim, G Michalopoulos… - Findings of the …, 2023 - aclanthology.org
Recent studies on automatic note generation have shown that doctors can save significant
amounts of time when using automatic clinical note generation (Knoll et al., 2022) …