Scholar
Niharika S. D'Souza
Google Scholar ID: iKonuvcAAAAJ
Research Scientist, IBM Research, Almaden
Citations & Impact
Citations
235
H-index
9
i10-index
9
Publications
20
Co-authors
32
list available
Publications
Resume (English only)
Academic Achievements
- Paper on 'Semantic Alignment in Vision Language Transformers' accepted into the workshop on Unifying Representations in Neural Models at NeurIPS 2025.
- Paper on 'Prompt Optimization without Task Cues and Instructions' accepted into the IEEE International Conference on Data Mining (ICDM) 2025, Demo Track.
- Paper on 'Phrasally Grounded Fact Checking for VLM based Automated Radiology Report Generation' accepted into MICCAI 2025.
- Paper on 'Prompt Optimization' accepted to VLDB'25 workshop on LLM+KG.
- New work on 'LLMs as Tabular Representation Learners' accepted to TMLR.
- Nominated for Full Membership into Sigma Xi, the Scientific Honor Society.
- Serving as an Area Chair for MICCAI 2025 and MIDL 2025.
- Released MMMG toolkit for multi-graph NNs.
- Work on 'Encoded Representations and Modern Hopfield Networks' accepted into the workshop on Unifying Representations in Neural Models at NeurIPS 2024 and selected for an oral spotlight.
- Work on 'Geometrically Constrained U-Nets for segmentation in Radial Imaging modalities' presented at the Machine Learning with Medical Imaging workshop at MICCAI 2024.
- Team recognized by IBM Research with an A-Level Technical Accomplishment for fundamental advances to the science of multimodal fusion.
- Recognized as an Outstanding Reviewer (one among the top 12 reviewers) for MICCAI 2023.
- Presented work on 'Maximal Correlation informed Multi-Layered GNNs for Multimodal Fusion' at the ML4MHD workshop at ICML 2023 as an oral.
- Contributed chapter 'Network Comparisons and their applications in Connectomics' appeared in 'Connectome Analysis: Characterization, Methods, and Analysis'.
- Served as a session chair for the session on Brain Connectomics at IPMI 2023.
- Presented work on 'Geodesic Mean Estimation for Functional Connectomics manifolds' at IPMI 2023 as an oral.
- Work from 2022 on 'multiplexed graph neural networks for multimodal fusion' recognized as a finalist for the Young Scientist Award for MICCAI 2022, and an NIH Travel Award.
- Recognized as one of the top 10% of reviewers for ICML 2022.
- During her doctoral studies, she developed a suite of mathematical models of brain and behavior spanning network optimization models, deep-generative hybrids, graph neural networks, and manifold learning approaches for analyzing functional and structural connectomics data.
Research Experience
- Since January 2022, she has been working as a research scientist at IBM Research, Almaden. Her research projects include semantic alignment, prompt optimization, fact checking, and multimodal fusion.
Education
- She obtained her doctoral degree from the Electrical and Computer Engineering at Johns Hopkins University between 2016-2021, under the supervision of Dr. Archana Venkataraman.
Background
- Her research interests are statistical representation learning, geometric deep learning, graph signal processing, medical computer vision, and tabular deep learning.
Miscellany
- Moderating a discussion on Scalable and Translatable Healthcare Solutions at the Conference on Health, Inference, and Learning (CHIL 2025) as a Senior Roundtable Leader; serving as an Organising Committee Member for the 6th workshop on GRaphs in biomedicAl Image anaLysis (GRAIL 2024).
