An animated picture says at least a thousand words: Selecting Gif-based Replies in Multimodal Dialog

Xingyao Wang, David Jurgens


Abstract
Online conversations include more than just text. Increasingly, image-based responses such as memes and animated gifs serve as culturally recognized and often humorous responses in conversation. However, while NLP has broadened to multimodal models, conversational dialog systems have largely focused only on generating text replies. Here, we introduce a new dataset of 1.56M text-gif conversation turns and introduce a new multimodal conversational model Pepe the King Prawn for selecting gif-based replies. We demonstrate that our model produces relevant and high-quality gif responses and, in a large randomized control trial of multiple models replying to real users, we show that our model replies with gifs that are significantly better received by the community.
Anthology ID:
2021.findings-emnlp.276
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2021
Month:
November
Year:
2021
Address:
Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
Findings
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
3228–3257
Language:
URL:
https://aclanthology.org/2021.findings-emnlp.276
DOI:
10.18653/v1/2021.findings-emnlp.276
Bibkey:
Cite (ACL):
Xingyao Wang and David Jurgens. 2021. An animated picture says at least a thousand words: Selecting Gif-based Replies in Multimodal Dialog. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 3228–3257, Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
An animated picture says at least a thousand words: Selecting Gif-based Replies in Multimodal Dialog (Wang & Jurgens, Findings 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.findings-emnlp.276.pdf
Video:
 https://aclanthology.org/2021.findings-emnlp.276.mp4
Code
 xingyaoww/gif-reply
Data
GIF Reply Dataset