Sandeep Kumar

sandeep.kumar82945 [at] gmail.com
  resume
  linkedin
  github
  scholar

Hello!

Hi! I’m Sandeep, a fourth-year Year Prime Minister Research Fellow (PMRF) at the AI-NLP-ML Research Lab, Department of Computer Science and Engineering, Indian Institute of Technology Patna. I am working with Prof. Asif Ekbal. Previously, I did my Masters from from Indian Institute of Technology Patna, India, where I was advised by Prof. Asif Ekbal.

My research interests lie in the intersection of Natural Language Processing (NLP) and Machine Learning (ML), with a focus on understanding and extending the capabilities of Large Language Models (LLMs). I am particularly interested in developing AI agents that can effectively communicate and cooperate with humans, as well as enhancing the evaluation methodologies for LLMs. My work also explores the application of LLMs to complex, document-level tasks and scientific texts, addressing the broader challenge of creating trustworthy machine learning systems that are secure, interpretable, and aligned with safety principles.

News and Updates

  • Jun 2025: Our work with Samsung Research, Bengaluru "CourtEval: A Courtroom-Based Multi-Agent Evaluation Framework" got accepted at ACL 2025.
  • Jun 2025: Our Workshop proposal "JUST-NLP: Workshop on NLP for Empowering Justice" got accepted at AACL-IJCNLP 2025 to be held in December 2025, Mumbai.
  • April 2025: I attended NAACL 2025 in Albuquerque, New Mexico, where I presented our paper, 'MixRevDetect: Towards Detecting AI-Generated Content in Hybrid Peer Reviews.' .
  • Nov 2024: I attended EMNLP 2024 in Miami, Florida, where I presented our paper, 'Quis custodiet ipsos custodes?'Who will watch the watchmen? On Detecting AI-generated peer-reviews
  • Sept 2024: Our Paper "Can Large Language Models Unlock Novel Scientific Research Ideas?" is published on Arxiv.
  • August 2024: I attended ACL 2024 in Bangkok, Thailand
  • May 2024: I attended LREC-COLING 2024 in Turin, Italy, where I presented our paper, Longform multimodal lay summarization of scientific papers: Towards automatically generating science blogs from research articles .
  • December 2023: I attended EMNLP 2023 in Singapore, where I presented our paper, When Reviewers Lock Horn: Finding Disagreement in Scientific Peer Reviews .
  • July 2021: Started my PhD at IIT Patna

Publications

MixRevDetect: Towards Detecting AI-Generated Content in Hybrid Peer Reviews.
Sandeep Kumar, S Garg, S Sengupta, T Ghosal, A Ekbal
NAACL (Main) '25 | The Nations of the Americas Chapter of the Association for Computational Linguistics 2025
paper

'Quis custodiet ipsos custodes?'Who will watch the watchmen? On Detecting AI-generated peer-reviews
Sandeep Kumar, Mohit Sahu, Vardhan Gacche, Tirthankar Ghosal, Asif Ekbal
EMNLP (Main) '24 | Conference on Empirical Methods in Natural Language Processing 2024
paper

Can Large Language Models Unlock Novel Scientific Research Ideas?
Sandeep Kumar, Tirthankar Ghosal, Vinayak Goyal, Asif Ekbal
ArXiv
paper

Longform multimodal lay summarization of scientific papers: Towards automatically generating science blogs from research articles
Sandeep Kumar, Guneet Singh Kohli, Tirthankar Ghosal, Asif Ekbal
LREC-COLING'24 | The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation 2024
paper slides

When Reviewers Lock Horn: Finding Disagreement in Scientific Peer Reviews
Sandeep Kumar, Tirthankar Ghosal, Asif Ekbal
EMNLP' (Main) 23 | The 2023 Conference on Empirical Methods in Natural Language Processing
paper code and dataset Poster



Invited Talks & Symposiums

2025|   |   Got invitation to attend and present our work at PMRF Symposium 2025 held at IIT Hyderabad.
Annual Symposium - 2025   |   Youtube

2025 | 2023   |   Got invitation to attend and present our work at Academic Research and Careers for Students (ARCS) is an annual symposium hosted by ACM India held on February 27-28 in Coimbatore
ACM India ARCS- 2025   |  


Position of Responsibility

  • PC member at AACL-IJCNLP 2025 -- to be held at Mumbai, India 2025 .
  • PC member of SIGDial 2021 Special Session on Summarizing of Dialogues and Multi-Party Meetings SIGDial 2021 .

Teaching

  • Artificial Intelligence (Spring 2022, Spring 2021)
  • C++ Programming (Spring 2023)
  • Computer Architecture (Autumn 2021)

  Template: Ashish Sharma