Shravan Venkatraman

I am an M.Sc. computer vision student at MBZUAI. I am fortunate to be a part of the Intellectual and Visual Analytics Lab, where I am advised by Dr. Fahad Khan and Dr. Salman Khan. I completed my undergrad in computer science at VIT University, advised by Dr. Joe Dhanith and Dr. Pandiyaraju (also receiving the Sir C. V. Raman Award three times during my time there).

My research focuses on developing self-evolving large multimodal models for generalizable multimodal intelligence, within the broader context of multimodal representation learning for reasoning. I also work on unified large-scale models for image understanding and generation, with a focus on geometry-aware representations and neural rendering in computer graphics.

Prior to this, I was a research intern at Nagasaki University advised by Dr. Muthu Subash Kavitha. In the summer of 2024, I interned at MedxAI under the mentorship of Dr. Susan Elias and Dr. Sheena Pravin. I also work on industry- and consultancy-funded projects through SPORIC at VIT.

I'm always happy to chat about research, collaborations, or startups — feel free to send me an email! :)

Email CV Scholar GitHub

Shravan Venkatraman

News

Jan 13, 2026 – I've been selected for the MBZUAI ML Winter School 2026 on Representation Learning & GenAI!
Dec 16, 2025 – Grateful to have received the Sir C. V. Raman Award from VIT Chennai for the third time (post-graduation, too) for my work on SAG-ViT.
Nov 21, 2025 – Our paper on EvoLMM, a purely self-evolving framework for LMMs, is now on arXiv!
Nov 27, 2025 – SPROUT has been accepted to Neurocomputing!
Oct 27, 2025 – RG-ViT has been accepted to Computers in Biology and Medicine!
Sep 18, 2025 – I'm honored to serve as a Student Representative for the Computer Vision department at MBZUAI!
Aug 10, 2025 – I'm excited to start my MSc. in Computer Vision at MBZUAI!
Jul 27, 2025 – SAG-ViT has been accepted to Complex and Intelligent Systems!
Jul 14, 2025 – UGPL is accepted to ICCV'25 Workshops: CVAMD! Paper and Code are available!
Apr 17, 2025 – I have successfully defended my bachelor's thesis (titled: Making NeRF See Structure, Not Just Light) at VIT Chennai!
Feb 28, 2025 – FUSION is accepted to CVPR'25 Workshops: NTIRE!
Apr 07, 2025 – Honored to receive the Sir C. V. Raman Award from VIT Chennai for the second time in recognition of my research!
Feb 28, 2025 – We showcased and presented CerviLens at IInvenTiv'25 @IIT Madras, representing MedxAI Innovations!
Jan 25, 2025 – I am honored to have been admitted to the MSc. Computer Vision program at MBZUAI!
Dec 12, 2024 – Proud to have been selected as a recipient of the Sir C. V. Raman Award by VIT Chennai for my research!
Jun 23, 2024 – I presented our paper on attention-fused deep CNNs at ICRAS 2024 in Tokyo, Japan!
Selected Publications

Hover over publications for quick preview

EvoLMM
EvoLMM: Self-Evolving Large Multimodal Models with Continuous Rewards
arXiv
EvoLMM
EvoLMM
EvoLMM: Self-Evolving Large Multimodal Models with Continuous Rewards
Omkar Thawakar*, Shravan Venkatraman*, Ritesh Thawkar*, Abdelrahman M Shaker, Hisham Cholakkal, Rao Muhammad Anwer, Salman Khan, Fahad Shahbaz Khan
arXiv
paper / code / project page / abs / bibtex

EvoLMM is a fully unsupervised self-evolving framework for large multimodal models (LMMs) that improves visual reasoning from raw images only by coupling a Proposer and a Solver trained via continuous self-consistency rewards.

PDE-NeRF
Making NeRF See Structure, Not Just Light: Enforcing PDE-Based Surface Constraints
Pattern Recognition
Making NeRF See Structure, Not Just Light: Enforcing PDE-Based Surface Constraints for 3D Consistency
Shravan Venkatraman, Pandiyaraju V
Pattern Recognition
code & paper: out soon

Enforcing physical surface properties through PDE constraints yields geometrically accurate neural scene representations from sparse views.

SAG-ViT
SAG-ViT: A Scale-Aware, High-Fidelity Patching Approach with Graph Attention
Complex and Intelligent Systems
SAG-ViT: A Scale-Aware, High-Fidelity Patching Approach with Graph Attention for Vision Transformers
Shravan Venkatraman, Jaskaran Singh Walia, Joe Dhanith P R
Complex and Intelligent Systems
code / paper / Hugging Face

Structuring attention through multi-scale graphs enable transformers to reason across visual hierarchies.

Teaching
dragon

BCSE332P - Deep Learning Lab (Fall 2024)


Last updated January 2026.