My current research focuses on LLM collaboration, evaluation, and agentic reasoning. I also conduct empirical, mixed-methods research in human-AI interaction and enjoy publishing at both AI and HCI venues that bridge the two communities.

Publications

Natural Language Processing, Machine Learning

nlp-bea18-thumbnail.png Lexical complexity-driven representation learning
(Master’s Thesis, Aug 2022)
Nikhil Wani
University of Southern California, Los Angeles
[pdf]

temp.png The Whole is Greater than the Sum of its Parts: Towards the Effectiveness of Voting Ensemble Classifiers for Complex Word Identification
(NAACL 2018 Shared Task)
Nikhil Wani, Sandeep Mathias, Jayashree Aanad Gajam, Pushpak Bhattacharyya
NLP-BEA@ NAACL 2018
[pdf]

@inproceedings{wani2018whole,
title={The Whole is Greater than the Sum of its Parts: Towards the Effectiveness of Voting Ensemble Classifiers for Complex Word Identification},
author={Wani, Nikhil and Mathias, Sandeep and Gajjam, Jayashree Aanand and Bhattacharyya, Pushpak},
booktitle={Proceedings of the Thirteenth Workshop on Innovative Use of NLP for Building Educational Applications},
pages={200--205},
year={2018}
}

Human-Computer Interaction, Human-AI Interaction

EA@CHI24

Large Language Models as Thought Partners: Collaborative Decision Making with LLMs
Nikhil Wani
CHI 2025, Case Study Track [Conditionally Accepted]; voluntarily withdrawn for a U.S. design patent.
[reviews]

Probabilistic Modeling of Swarachakra Mobile Keyboard for Improved Touch Accuracy
Nikhil Wani, Sumit Singh Yadav, Anirudha Joshi
INTERACT 2017, SRC Track. In the Adjunct Proceedings of 16th IFIP International Conference on Human-Computer Interaction
[pdf] [bib]

Pre-Prints / Technical Reports / Work-In-Progress

  • Is your Statement Purposeless? Predicting Computer Science Graduate Admission Acceptance based on Statement Of Purpose
    †Diptesh Kanojia, †Nikhil Wani, Pushpak Bhattacharyya | †= Equal contribution
    ICON 2017. In 15th International Conference on Natural Language Processing
    [pdf]

@InProceedings{kanojia-wani-bhattacharyya:2017:W17-75,
author = {Kanojia, Diptesh and Wani, Nikhil and Bhattacharyya, Pushpak},
title = {Is your Statement Purposeless? Predicting Computer Science Graduation Admission Acceptance based on Statement Of Purpose},
booktitle = {Proceedings of the 14th International Conference on Natural Language Processing (ICON-2017)},
month = {December},
year = {2017},
address = {Kolkata, India},
publisher = {NLP Association of India},
pages = {141--145},
url = {http://www.aclweb.org/anthology/W/W17/W17-7518}
}

  • Conducting Contextual Inquiry of Twitter for Work Engagement
    Nikhil Wani, Ganesh Bhutkar
    IJCA 2017. In International Journal of Computer Applications
    [pdf] [bib]

Program Committee / Reviewer / Academic Services

  • Teaching & other academic services:
    • University of Southern California
      • Graduate courses:
      • CSCI 570: Analysis of Algorithms
        • Course Producer (Fall 2020, Summer 2021): Prof. Shahriar Shamsian and Prof. Victor Adamchik.
      • CSCI 585: Database systems
        • Grader (Spring 2021) : Prof. Saty Raghavachar
      • CSCI 571: Web Technologies (CS Honors Program assignment)
        • Head TA (Fall 21, Spring 22, and Summer 22): Prof. Marco Papa

Research Implementation

  • NLP Tasks:
    • Refer to my recent Pytorch + HuggingFace implementation demos page [In-progress].