Silpa Vadakkeeveetil Sreelatha

PhD Student

s.vadakkeeveetilsreelatha [AT] surrey.ac.uk

Bio

I am an ELLIS Ph.D. student specializing in Machine Learning and Computer Vision, co-advised by Dr. Anjan Dutta (University of Surrey), Prof. Serge Belongie (University of Copenhagen), and Dr. Muhammad Rana (University of Surrey). I am fortunate to be affiliated with the Surrey Institute for People-Centred Artificial Intelligence (PAI), Centre for Vision, Speech and Signal Processing (CVSSP) at the University of Surrey, and Pioneer Centre for Artificial Intelligence in Copenhagen. Before beginning my Ph.D., I worked as a full-time researcher (2019-2022) at Tata Research and Innovation Labs, focusing on Deep Learning and AI under the guidance of Principal Scientist Dr. Shirish Subhash Karande.

My goal is to discover interpretable representations within deep learning models to advance fairness, robustness, and transparency in both discriminative and generative models. I am passionate about exploring how interpretability can drive practical and ethical applications of AI across various domains.

When I am Not Busy Being a Research Wizard...

I can usually be found singing along to my favorite songs or getting lost in a good movie. I love catching up with friends, sharing laughs, and enjoying the little moments that make life special!

I am always open to collaborations. If you're interested in working together, feel free to email me!

News

06/2024: Our paper on Responsible T2I generation accepted in SafeGen AI workshop, NeurIPS2024!
09/2024: Our paper on Debiasing accepted in NeurIPS2024. See you in Vancouver!
06/2023: Excited to start as a PhD student at the University of Surrey.
12/2022: Shortlisted for the prestigious ELLIS PhD program!
11/2021: Our paper on Interpretable latent discovery in GANs accepted in AAAI'2022.
04/2020: Our paper on Advertisement undertsanding accepted in ACL'2022.
04/2021: Our work on Disentanglement based Active Learning was accepted at IJCNN'2021.
12/2020: Thrilled to attend my first-ever in-person conference - and it's NeurIPS 2020!
04/2020: Our work on Disentanglement based Active Learning accepted at CtrlGen workshop, NeurIPS2020.
06/2019: Started working as Deep Learning Reseacher at TCS Research.

Publications & Patents

indicates equal contribution.

  • Papers
  • Patents

Concept Denoising Score Matching for Responsible Text-to-Image Generation

Silpa Vadakkeeveetil Sreelatha, Sauradip Nag, Serge Belongie, Muhammad Awais, Anjan Dutta

Safe Generative AI Workshop, NeurIPS 2024

DeNetDM: Debiasing by Network Depth Modulation

Silpa Vadakkeeveetil Sreelatha, Adarsh Kappiyath, Abhra Chaudhuri, Anjan Dutta

NeurIPS'24

Self-Supervised Enhancement of Latent Discovery in GANs

Adarsh Kappiyath, Silpa Vadakkeeveetil Sreelatha, S Sumitra

AAAI'22

Understanding Advertisements with BERT

Kanika Kalra, Bhargav Kurma, Silpa Vadakkeeveetil Sreelatha, Manasi Patwardhan, Shirish Karande

ACL'22

Disentanglement based Active Learning

Adarsh Kappiyath, Silpa Vadakkeeveetil Sreelatha, S Sumitra

IJCNN'21

Method and system for visio-linguistic understanding using contextual language model reasoners

Sai Sree Bhargav Kurma, Kanika Kalra, Silpa Vadakkeeveetil Sreelatha, Manasi Patwardhan, Shirish Subhash Karande

US Patent

Academic Service

Conference Reviewing: NeurIPS'2023, NeurIPS'2024, CVPR'2024, ICML'2024

Workshop Reviewing: WiCV CVPR'2023, SafeGenAI NeurIPS'2024

Vitæ

Full Resume in PDF.

  • University of Surrey June 2023 - now
    Ph.D. Student
    People-Centred Artificial Intelligence
  • TCS Research July 2019-April 2022
    Researcher
    Deep Learning & AI
  • Indian Institute of Space Science and Technology, India July 2017 – June 2019
    M.Tech. Student
    Machine Learning and Computing
  • Government Engineering College, Sreekrishnapuram Aug 2012 - May 2016
    B.Tech Student
    Computer Science and Engineering

Website Design

Huge thanks to Prof. Martin Saveski for providing this website theme. The code needed to build the website is in this GitHub repo. Also, a big shoutout to Sruthi Aduvattil for using her creativity to help me fine-tune the original theme and bring my not-so-creative suggestions to life!