6 papers
Who We Are, Where We Are: Mental Health at the Intersection of Person, Situation, and Large Language Models
Nikita Soni, August HÃ¥kan Nilsson, Syeda Mahwish +3
Mental health is not a fixed trait but a dynamic process shaped by the interplay between individual dispositions and situational contexts. Building on interactionist and constructi…
Addressing the Ecological Fallacy in Larger LMs with Human Context
Nikita Soni, Dhruv Vijay Kunjadiya, Pratham Piyush Shah +3
Language model training and inference ignore a fundamental linguistic fact -- there is a dependence between multiple sequences of text written by the same person. Prior work has sh…
Residualized Similarity for Faithfully Explainable Authorship Verification
Peter Zeng, Pegah Alipoormolabashi, Jihu Mun +5
Responsible use of Authorship Verification (AV) systems not only requires high accuracy but also interpretable solutions. More importantly, for systems to be used to make decisions…
Evaluation of LLMs-based Hidden States as Author Representations for Psychological Human-Centered NLP Tasks
Nikita Soni, Pranav Chitale, Khushboo Singh +2
Like most of NLP, models for human-centered NLP tasks -- tasks attempting to assess author-level information -- predominantly use representations derived from hidden states of Tran…
Comparing Pre-trained Human Language Models: Is it Better with Human Context as Groups, Individual Traits, or Both?
Nikita Soni, Niranjan Balasubramanian, H. Andrew Schwartz +1
Pre-trained language models consider the context of neighboring words and documents but lack any author context of the human generating the text. However, language depends on the a…
Large Human Language Models: A Need and the Challenges
Nikita Soni, H. Andrew Schwartz, João Sedoc +1
As research in human-centered NLP advances, there is a growing recognition of the importance of incorporating human and social factors into NLP models. At the same time, our NLP sy…