Transformers Go for the LOLs: Generating (Humourous) Titles from Scientific Abstracts End-to-End

Yanran Chen, Steffen Eger


Abstract
We consider the end-to-end abstract-to-title generation problem, exploring seven recent transformer based models (including ChatGPT) fine-tuned on more than 30k abstract-title pairs from NLP and machine learning (ML) venues. As an extension, we also consider the harder problem of generating humorous paper titles. For the latter, we compile the first large-scale humor annotated dataset for scientific papers in the NLP/ML domains, comprising 2.6k titles. We evaluate all models using human and automatic metrics. Our human evaluation suggests that our best end-to-end system per-forms similarly to human authors (but arguably slightly worse). Generating funny titles is more difficult, however, and our automatic systems clearly underperform relative to humans and often learn dataset artefacts of humor. Finally, ChatGPT, without any fine-tuning, performs on the level of our best fine-tuned system.
Anthology ID:
2023.eval4nlp-1.6
Volume:
Proceedings of the 4th Workshop on Evaluation and Comparison of NLP Systems
Month:
November
Year:
2023
Address:
Bali, Indonesia
Editors:
Daniel Deutsch, Rotem Dror, Steffen Eger, Yang Gao, Christoph Leiter, Juri Opitz, Andreas Rücklé
Venues:
Eval4NLP | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
62–84
Language:
URL:
https://aclanthology.org/2023.eval4nlp-1.6
DOI:
10.18653/v1/2023.eval4nlp-1.6
Bibkey:
Cite (ACL):
Yanran Chen and Steffen Eger. 2023. Transformers Go for the LOLs: Generating (Humourous) Titles from Scientific Abstracts End-to-End. In Proceedings of the 4th Workshop on Evaluation and Comparison of NLP Systems, pages 62–84, Bali, Indonesia. Association for Computational Linguistics.
Cite (Informal):
Transformers Go for the LOLs: Generating (Humourous) Titles from Scientific Abstracts End-to-End (Chen & Eger, Eval4NLP-WS 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.eval4nlp-1.6.pdf