NTULM: Enriching Social Media Text Representations with Non-Textual Units

Jinning Li, Shubhanshu Mishra, Ahmed El-Kishky, Sneha Mehta, Vivek Kulkarni


Abstract
On social media, additional context is often present in the form of annotations and meta-data such as the post’s author, mentions, Hashtags, and hyperlinks. We refer to these annotations as Non-Textual Units (NTUs). We posit that NTUs provide social context beyond their textual semantics and leveraging these units can enrich social media text representations. In this work we construct an NTU-centric social heterogeneous network to co-embed NTUs. We then principally integrate these NTU embeddings into a large pretrained language model by fine-tuning with these additional units. This adds context to noisy short-text social media. Experiments show that utilizing NTU-augmented text representations significantly outperforms existing text-only baselines by 2-5% relative points on many downstream tasks highlighting the importance of context to social media NLP. We also highlight that including NTU context into the initial layers of language model alongside text is better than using it after the text embedding is generated. Our work leads to the generation of holistic general purpose social media content embedding.
Anthology ID:
2022.wnut-1.7
Volume:
Proceedings of the Eighth Workshop on Noisy User-generated Text (W-NUT 2022)
Month:
October
Year:
2022
Address:
Gyeongju, Republic of Korea
Venue:
WNUT
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
69–82
Language:
URL:
https://aclanthology.org/2022.wnut-1.7
DOI:
Bibkey:
Cite (ACL):
Jinning Li, Shubhanshu Mishra, Ahmed El-Kishky, Sneha Mehta, and Vivek Kulkarni. 2022. NTULM: Enriching Social Media Text Representations with Non-Textual Units. In Proceedings of the Eighth Workshop on Noisy User-generated Text (W-NUT 2022), pages 69–82, Gyeongju, Republic of Korea. Association for Computational Linguistics.
Cite (Informal):
NTULM: Enriching Social Media Text Representations with Non-Textual Units (Li et al., WNUT 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.wnut-1.7.pdf
Data
TweetEval