Umakanta Maharana

PhD Scholar in Computer Vision @ NIT Rourkela

Ph.D. Scholar (Computer Vision)

NIT Rourkela, Odisha

526cs6003@nitrkl.ac.in

Ph.D. Scholar @ NIT Rourkela • ANRF PAIR Project

Hi! I am Umakanta Maharana, a doctoral researcher pursuing my Ph.D. in Computer Science & Engineering at the National Institute of Technology (NIT) Rourkela, specializing in Computer Vision and Deep Learning.

My doctoral research investigates Cattle Behaviour Recognition under the prestigious ANRF PAIR (Partnerships for Accelerated Innovation and Research) project. Under this flagship national initiative—where NIT Rourkela serves as a national hub institution—my research focuses on designing deep spatio-temporal video architectures, action recognition models, and vision-based precision livestock monitoring systems for automated animal welfare, activity tracking, and health diagnostics.

Prior to starting my Ph.D., I was a Research Fellow at the RespAI Lab, where I worked on Role-Based Access Control in LLMs (including our OrgAccess benchmark, accepted at EMNLP 2025), healthcare AI diagnostic reasoning faithfulness, and selective machine unlearning under a SERB Government of India research grant.

Core Research Interests

Computer Vision Cattle & Livestock Analytics Video Action Recognition Deep Learning AI Safety & Alignment Role-Based Access Control Machine Unlearning

526cs6003@nitrkl.ac.in Personal Email Scholar GitHub LinkedIn Twitter

Updates

All updates

Aug 01, 2026	🏛️ Commenced my Ph.D. in Computer Science & Engineering at the National Institute of Technology (NIT) Rourkela! My research centers on Computer Vision for Cattle Behaviour Recognition under the prestigious ANRF PAIR (Partnerships for Accelerated Innovation and Research) project.
Aug 20, 2025	🎉 Two papers accepted at EMNLP 2025! Excited to share that “OrgAccess: A Benchmark for Role-Based Access Control in Organization-Scale LLMs” and “Investigating Pedagogical Teacher and Student LLM Agents” have both been accepted at EMNLP 2025.
Apr 10, 2025	Preprint of “Right Prediction, Wrong Reasoning: Uncovering LLM Misalignment in RA Disease Diagnosis” is available on Arxiv.

Latest Research Blogs

All blogs

May 27, 2025	Can AI Respect Company Secrets? Introducing OrgAccess
May 26, 2025	Teaching AI How to Teach: Simulating Classrooms with Adaptive Agents
Apr 11, 2025	Right Prediction, Wrong Reasoning: Why High Accuracy in Medical AI Can Be Dangerous

Selected Publications

All publications

Preprint

Right Prediction, Wrong Reasoning: Uncovering LLM Misalignment in RA Disease Diagnosis

Umakanta Maharana, Sarthak Verma, Avarna Agarwal, and 4 more authors

2025

Under review. Investigating the implications of LLM misalignment and reasoning faithfulness in medical diagnostics.

Abs arXiv Bib PDF Blog

Large Language Models (LLMs) frequently achieve high classification accuracy on clinical diagnosis benchmarks while relying on medically flawed or hallucinated reasoning steps. This study evaluates diagnostic justifications generated by LLMs for Rheumatoid Arthritis (RA) against blinded clinical rheumatology experts, demonstrating that high benchmark accuracy can be deceptive and dangerous in real-world clinical deployment.

@misc{maharana2025rightpredictionwrongreasoning,
  title = {Right Prediction, Wrong Reasoning: Uncovering LLM Misalignment in RA Disease Diagnosis},
  author = {Maharana, Umakanta and Verma, Sarthak and Agarwal, Avarna and Mruthyunjaya, Prakashini and Mahapatra, Dwarikanath and Ahmed, Sakir and Mandal, Murari},
  year = {2025},
  eprint = {2504.06581},
  archiveprefix = {arXiv},
  primaryclass = {cs.AI},
  url = {https://arxiv.org/abs/2504.06581},
  note = {Under review. Investigating the implications of LLM misalignment and reasoning faithfulness in medical diagnostics.},
  keywords = {LLM misalignment, RA disease diagnosis, medical diagnostics, AI ethics, machine learning},
  journal = {arXiv preprint arXiv:2504.06581},
}

EMNLP

OrgAccess: A Benchmark for Role-Based Access Control in Organization-Scale LLMs

Debdeep Sanyal, Umakanta Maharana, Hong Ming Tan, and 4 more authors

In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP) , 2025

Accepted at EMNLP 2025. A benchmark evaluating whether models can reliably respect organizational roles, permissions, and access boundaries.

Abs arXiv Bib PDF Blog Code Hugging Face

We introduce OrgAccess, a benchmark for evaluating role-based access control in organization-scale LLMs, focusing on whether models can reliably respect organizational roles, permissions, and access boundaries. OrgAccess puts 16 frontier and open models against real enterprise-style access rules — 40 NIST-grounded permission types, 70K instances, up to 5 conflicting constraints at once.

@inproceedings{sanyal2025orgaccess,
  title = {OrgAccess: A Benchmark for Role-Based Access Control in Organization-Scale LLMs},
  author = {Sanyal, Debdeep and Maharana, Umakanta and Tan, Hong Ming and Sinha, Yash and Karande, Shirish and Kankanhalli, Mohan and Kumar, Dhruv},
  booktitle = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  year = {2025},
  eprint = {2505.19165},
  archiveprefix = {arXiv},
  primaryclass = {cs.AI},
  url = {https://arxiv.org/abs/2505.19165},
  note = {Accepted at EMNLP 2025. A benchmark evaluating whether models can reliably respect organizational roles, permissions, and access boundaries.},
  keywords = {Large Language Models, Role-Based Access Control, AI Safety, Benchmark, Enterprise AI},
}

EMNLP
Investigating Pedagogical Teacher and Student LLM Agents: Genetic Adaptation Meets Retrieval Augmented Generation Across Learning Style

Debdeep Sanyal, Agniva Maiti, Umakanta Maharana, and 4 more authors

In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP) , 2025

Accepted at EMNLP 2025. An adaptive framework integrating heterogeneous student agents with genetic teacher policy evolution.

Abs arXiv Bib PDF Blog

Effective teaching requires adapting instructional strategies to accommodate the diverse cognitive and behavioral profiles of students, a persistent challenge in education and teacher training. We introduce a novel simulation framework that integrates LLM-based heterogeneous student agents with a self-optimizing teacher agent whose pedagogical policy dynamically evolves using a genetic algorithm. In addition, we propose Persona-RAG, a Retrieval Augmented Generation module that enables student agents to retrieve knowledge tailored to their individual learning styles.
@inproceedings{sanyal2025pedagogical, title = {Investigating Pedagogical Teacher and Student LLM Agents: Genetic Adaptation Meets Retrieval Augmented Generation Across Learning Style}, author = {Sanyal, Debdeep and Maiti, Agniva and Maharana, Umakanta and Kumar, Dhruv and Mali, Ankur and Giles, C. Lee and Mandal, Murari}, booktitle = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP)}, year = {2025}, eprint = {2505.19173}, archiveprefix = {arXiv}, primaryclass = {cs.AI}, url = {https://arxiv.org/abs/2505.19173}, note = {Accepted at EMNLP 2025. An adaptive framework integrating heterogeneous student agents with genetic teacher policy evolution.}, keywords = {Pedagogical Agents, LLM, Multi-Agent Simulation, Retrieval-Augmented Generation, Genetic Algorithm}, }

Core Research Interests

Updates

Latest Research Blogs

Selected Publications

Connect With Me