Research
My research focuses on controlled text generation, user modeling, and personalization, leveraging conversational AI, retrieval-augmented generation (RAG), and multi-agent systems. My specialization is in large language model (LLM) alignment and fine-tuning, with experience applying techniques such as RLHF, DPO, and GRPO alongside prompt chaining. My work is applied across diverse domains, including social media, code generation, education, and creative writing.
Google Scholar, Semantic Scholar
Below is a list of my publications.
* denotes equal contribution.
Conferences and Workshops
Whose story is it? Personalizing story generation by inferring author styles
N. Ashok Kumar, C.M. Pham, M. Iyyer, A. Lan
AACL-IJCNLP 2025
Code + Data
PRACTIQ: A Practical Conversational text-to-SQL dataset with Ambiguous and Unanswerable Queries
N. Ashok Kumar*, M. Dong*, Y. Hu, A. Chauhan, C. Hang, S. Chang, L. Pan, W. Lan, H. Zhu, J. Jiang, P. Ng, Z. Wang
NAACL 2025
Code + Data
Improving Socratic Question Generation using Data Augmentation and Preference Optimization
N. Ashok Kumar, A. Lan
BEA Workshop @ NAACL 2024
Code
Using Large Language Models for Student-Code Guided Test Case Generation in Computer Science Education
N. Ashok Kumar, A. Lan
AI4ED Workshop @ AAAI 2024 ( Best Runner-Up Paper 🏆)
Code
Improving Reading Comprehension Question Generation with Data Augmentation and Overgenerate-and-rank
N. Ashok Kumar, N. Fernandez, Z. Wang, A. Lan
BEA Workshop @ ACL 2023 (Outstanding Paper 🏆)
Code
A Conceptual Model for End-to-End Causal Discovery in Knowledge Tracing
N. Ashok Kumar, W. Feng, J. Lee, H. McNichols, A. Ghosh, A. S. Lan
EDM 2023
Code
MMM: An Emotion and Novelty-aware Approach for Multilingual Multimodal Misinformation Detection
V.Gupta, R.Kumari, N.Ashok, T.Ghosal, A.Ekbal
Findings of AACL-IJCNLP 2022
Code
Testing the Generalization of Neural Language Models for COVID-19 Misinformation Detection
N.Ashok*, J.P.Wahle*, T.Ruas, N.Meuschke, T.Ghosal, B.Gipp
iConference 2022
Code
A Multitask learning approach for fake news detection: Novelty, Emotion and Sentiment Lend a Helping Hand
N.Ashok*, R.Kumari*, T.Ghosal, A.Ekbal
IJCNN 2021
Code
Journals
Misinformation Detection using Multi-Task Learning with Mutual Learning for Novelty and Emotion
N.Ashok*, R.Kumari*, T.Ghosal, A.Ekbal
Information Processing and Management, 2021
Code
What the Fake? Probing Misinformation Detection Standing on the Shoulder of Novelty and Emotion
N.Ashok*, R.Kumari*, T.Ghosal, A.Ekbal
Information Processing and Management, 2022
Code
Identifying Multimodal Misinformation Leveraging Novelty Detection and Emotion Recognition
N.Ashok*, R.Kumari*, P.K. Agarwal, T.Ghosal, A.Ekbal
Journal of Intelligent Information Systems, 2023
Code
Emotion aided multi-task framework for video embedded misinformation detection
R.Kumari, V.Gupta, N.Ashok, T.Ghosal, A.Ekbal
Multimedia Tools and Applications, 2023
Pre-Prints
Toward LLM-Supported Automated Assessment of Critical Thinking Subskills
N. Ashok Kumar*, M.C. Peczuh*, R. Baker, B. Lehman, D. Eisenberg, C. Mills, K. Chebrolu, S. Nashi, C. Young, B. Liu, S. Lachman, A. Lan
arXiv, 2025