About Me

I am a final-year Computer Science undergraduate at Nanyang Technological University (NTU), Singapore, thankful to be supervised by Dr. Chng Eng Siong and Dr. Rao Wei on my FYP in avian acoustic monitoring. I am also currently working with Dr. Alvin Guo Wei Chan as Research Assistant on large language model hallucinations and creativity.

At Panasonic R&D Singapore, under the mentorship of Srikanth Nagisetty, I developed ReCoOP, a stereo SELD framework that placed 3rd globally in the IEEE AASP DCASE 2025 Challenge. In the MBZUAI UGRIP 2025 program, I worked with Dr. Mohammad Yaqub, Dr. Hu Wang, and Dr. Mostafa Salem on MISFIT, a wavelet-based MRI modality synthesis framework and an end-to-end neuro-oncology pipeline, published in the MICCAI 2025 BraTS Challenge.

Earlier, with Dr. Christopher Ho Tin Lee and Zhao Chenhao, I worked on motor drive control, presenting our ADRC-neural network compensator for PMSMs at the International Conference of Undergraduate Research (ICUR) 2024.

These experiences have shaped my view that reliability and creativity in AI should go hand in hand, and I look forward to exploring new paths at this intersection.

Latest News

• Jan 2026
  Our paper received the Outstanding Paper Award at the AI4Research Workshop (AAAI 2026)
• Jan 2026
  I will be giving an oral presentation at the AI4Research Workshop (AAAI 2026) on 26 January 2026 for our paper, “Does Less Hallucination Mean Less Creativity? An Empirical Investigation in LLMs”

Publications

• AAAI 2026 Workshop on AI for Scientific Research

  Does less hallucination mean less creativity? An empirical investigation in LLMs.

  Oral presentation · Outstanding Paper Award

  Banerjee M., Wangsajaya, N. Y., Alsagoff, S. A. R., Tan, M. S., Choy, Z. K. C., Chan, A.
• Proceedings of the Workshop on Detection and Classification of Acoustic Scenes and Events (DCASE 2025)

  Exploiting Stereo Spatial Properties with ReCoOP Framework for joint Sound Event Detection and Localization.

  Banerjee M., S. Nagisetty, H. B. Teo
• Proceedings of the MICCAI 2025 BraTS Challenge

  MISFIT: Modality Inference via Style Fusion and Invertible Translation for Cross-Modality Synthesis of 3D MRI Volumes

  Banerjee M., Q. Ji, S. Hashmi, et al.
• Proceedings of the MICCAI 2025 BraTS Challenge

  EMedNeXt: An Enhanced Brain Tumor Segmentation Framework for Sub-Saharan Africa using MedNeXt V2 with Deep Supervision.

  A. Jaheen, A. Elsayed, Kim, D., Banerjee M., et al.
• Preprint

  Predicting Brain Tumor Response to Therapy using a Hybrid Deep Learning and Radiomics Approach

  D. Tikhonov, M. F. Scatolin, Banerjee M. et al.
• Proceedings of the URECA@NTU 2024-25

  Fact or Fiction: Can Truth Kill Creativity?

  Banerjee M., Alsagoff, S. A. R., A. G. W. Chan
• Workshop on Reasoning and Planning for Large Language Models, International Conference on Learning Representations (ICLR 2025)

  Automating Evaluation of Creativity in LLMs with Semantic Entropy and Efficient Multi-Agent Judge

  M. S. Tan, Z. Choy, Banerjee M. et al.
• Proceedings of the International Conference on Distributed Artificial Intelligence (DAI 2024), Student Poster Track

  SOPPU: Scalable One PEFT per User

  Banerjee M., Y. Jain
• Preprint

  Enhancing Urban Mobility through Adaptive Traffic Analysis: A Case Study in Singapore

  Banerjee M., A. C. Somasundaram, R. Pahwa, et al.
• Proceedings of the URECA@NTU 2023-24 (Orally presented at the International Conference of Undergraduate Research (ICUR 2024))

  Advanced Control Strategy for Motor Drives Using Neural Network Algorithm

  Banerjee M., C. Zhao, C. H. T. Lee