Prior to this, I was a Research Associate at Adobe, Media and Data Science Research Lab where I spent about 2 years focussing on computer vision (multimodal intelligence, image editing, style transfer) and language modeling (bias, privacy). Even before that, I did my undergraduate studies at Delhi Technological University (DCE) in Computer Science and Engineering. During my undergrad, I had the good fortune to colaborate with Camera Culture Group (MIT Media Lab) on privacy preserving computer vision, with Adobe on document intelligence, and with IIT Bombay on low-resource OCR systems.
I am broadly interested in multimodal machine learning with a focus on making adaptable (minimal supervision), ethical (fair and private), and practical (robust and efficient) systems for the real-world. I am applying for PhD positions for Fall 2025!
For more details, check my CV. The best way to reach me is via email.
Updates
Nov 2024: Preprint on unlearnable text datasets is out!
Nov 2024: Can you edit images using examples? Check our latest WACV 2025 paper, where we introduce ReEdit, a technique to make exemplar-based editing significanly better and faster.
Sept 2024: A timely work on LLM debiasing using prompts accepted at EMNLP (Main) 2024! Kudos to an amazing collaboration between Adobe MDSR and NUS!
Aug 2024: Joined Microsoft Research India as a Research Fellow. Looking forward to making LLMs more reliable and efficient!
Feb 2024: Presenting CAFIE at @AAAI 2024. I will also be giving a talk on the efficacy of prompting for debiasing LLMs, at the Student Abstract Session. Say hi in Vancouver BC!
Jul 2023: Giving an internship talk at DTU, excited to learn more about undergraduate research and new initiatives!
June 2023: Promoted to Research Associate 2!
May 2023: Paper accepted at EsFoMO@ ICML 2023! Learn how to mitigate biases during Knowledge Distillation using our latest method BIRD
Jan 2023: Paper accepted at WACV 2023! Find your document snippet in just one shot using MONOMER
October 2022: Presenting CBNS at ECCV 2022 a new way to protect point clouds from adversarial attacks during perception. This will be my first Conference, looking forward to make new friends in Tel Aviv!
Aug 2022: Joined Adobe Media and Data Science Research Lab as a Research Associate
Jul 2022: Graduated from Delhi Technological University with a Bachelor's in Computer Science and Engineering
Publications
Towards Operationalizing Right to Protect Data
, Simra Shahid*, and Chirag Agarwal
ArXiv'24 | Preprint (Under Review)
pdf
abstract
cite
ReEdit: Multimodal Exemplar based Image Editing with Diffusion Models
Ashutosh Srivastava*, Tarun Ram Menta*, , Silky Singh, Surgan Jandial,
and Balaji Krishnamurthy
WACV'25 | Winter Conference on Applications of Computer Vision
pdf
abstract
cite
Thinking Fair and Slow: On the Efficacy of Structured Prompts for Debiasing Language Models
Shaz Furniturevala*, Surgan Jandial*, , Pragyan Banerjee, Simra Shahid, Sumit Bhatia, and Kokil Jaidka
EMNLP'24 (Main) | Empirical Methods in Natural Language Processing
abstract
pdf|
cite
All Should be Equal in the Eyes of LMs
Pragyan Banerjee*, , Surgan Jandial*, Simra Shahid*, Shaz Furniturevala, Balaji Krishnamurthy, and Sumit Bhatia
AAAI'24 | The Association for the Advancement of Artificial Intelligence
abstract
pdf|
cite
Towards Fair Knowledge Distillation with Student Feedback
, Surgan Jandial*, and Chirag Agarwal
ICML'23 EsFoMo | Efficient Systems for Foundation Models Workshop@ ICML
abstract
pdf|
cite
Gatha: Relational Loss for enhancing text-based style transfer
Surgan Jandial, Shripad Deshmukh, , Simra Shahid, and Balaji Krishnamurthy
CVPR23'23 CV4AD | Computer Vision for Fashion, Art, and Design Workshop@ CVPR
abstract
pdf|
cite
One Shot Doc Snippet Detection: Powering Search in Document Beyond Text
, Shripad Deshmukh, Milan Agarwal, Surgan Jandial, Mausoom Sarkar, and Balaji Krishnamurthy
WACV'23 | Winter Conference on Applications of Computer Vision
abstract
pdf|
cite
Adaptive Split Learning
Ayush Chopra, Surya Kant Sahu, Abhishek Singh, , Praneeth Vepakomma, Mohammad Mohammadi Amiri, and Ramesh Raskar
FLSys' 23 | Federated Learning Systems (FLSys) Workshop@ MLSys
abstract
pdf|
cite
Learning to Censor by Noisy Sampling
Ayush Chopra, , Abhishek Singh, Vivek Sharma, and Ramesh Raskar
ECCV'22 | European Conference on Computer Vision
abstract
pdf|
cite
Rethinking Beford's Law with Neural Networks
Surya Kant Sahu, , Arshan Shaikh, and Yannic Kilcher
NeurIPS ML4PS'21 | Machine Learning and the Physical Sciences Workshop@ NeurIPS
abstract
pdf|
cite