Human Language Modeling
Nikita Soni, Matthew Matero, Niranjan Balasubramanian, H. Andrew Schwartz
Abstract
Natural language is generated by people, yet traditional language modeling views words or documents as if generated independently. Here, we propose human language modeling (HuLM), a hierarchical extension to the language modeling problem where by a human- level exists to connect sequences of documents (e.g. social media messages) and capture the notion that human language is moderated by changing human states. We introduce, HaRT, a large-scale transformer model for solving HuLM, pre-trained on approximately 100,000 social media users, and demonstrate it’s effectiveness in terms of both language modeling (perplexity) for social media and fine-tuning for 4 downstream tasks spanning document- and user-levels. Results on all tasks meet or surpass the current state-of-the-art.
BibTeX
@inproceedings{soni-etal-2022-human,
title = "Human Language Modeling",
author = "Soni, Nikita and
Matero, Matthew and
Balasubramanian, Niranjan and
Schwartz, H. Andrew",
editor = "Muresan, Smaranda and
Nakov, Preslav and
Villavicencio, Aline",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2022",
month = may,
year = "2022",
address = "Dublin, Ireland",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-acl.52/",
doi = "10.18653/v1/2022.findings-acl.52",
pages = "622--636"
}