Research

A curated collection of my research, organised by track.

Alignment & Safety in T2I and T2V Models

Designed SCFM, a training-free safety-alignment method for pretrained T2I diffusion models that steers denoising toward safer generations at inference time via safety-potential-guided rectified flow matching in frozen CLIP embedding space

TRCE Paper PresentationRead →
CSD 722: Depth Conditioned Video GenerationRead →
2 resourcesView Track →

Safety, Interpretability & Theory of LLMs

I worked on some disjoint topics around this broad track.

  • Attribution techniques for Interpretability
  • Enhancing Trust in LLMs
  • Transformers for Non-parametric Regression
Enhancing trust in LLMs notesRead →
Explainable AI: Attribution TechniquesRead →
Theory of LLMs — Notion NotesRead →
3 resourcesView Track →

UG Thesis

My undergraduate thesis on generative models for unsupervised speech time-scale modification, under the joint supervision of Prof. Prasanta Kumar Ghosh (SPIRE Lab, IISc) and Prof. Niteesh Sahni (SNIoE) — won 2nd Prize for Best UG Thesis.

Thesis PosterRead →
Speech Time Scale Modification with GANsRead →
Speech TSM using GANs - PresentationRead →
Statistics for Generative ModelsRead →
Denoising Diffusion Probabilistic Models NotesRead →
Generative Models: A Mathematical OverviewRead →
6 resourcesView Track →

PhD Research Courses

I completed 5 PhD-level courses during my undergraduate degree. These are the resources and notes I created while completing them.

  1. Advanced Deep Learning
  2. Advanced Computer Vision
  3. Special Topics in AI
  4. Measure & Integration
  5. Stochastic Processes
Latent Diffusion Model Paper PresentationRead →
CSD 722: Depth Conditioned Video GenerationRead →
Contrastive Learning: SimCLR & I-JEPARead →
Variational Autoencoders (VAEs)Read →
Vision Transformer (ViT)Read →
Statistics for Generative ModelsRead →
Lecture on Flow MatchingRead →
Denoising Diffusion Probabilistic Models NotesRead →
8 resourcesView Track →

Seminar and Lecture Notes

From my UG Seminar course, where we were taught how to write reports, papers, and give presentations — these are the resources and notes from that coursework.

Principal Component AnalysisRead →
Cross Validation TechniquesRead →
Sequential Models: RNNs OverviewRead →
Neural Networks: Foundations and ArchitecturesRead →
4 resourcesView Track →