Sriram Mandalika
Dual Studium SAP x HPI

Hello! I am currently an Applied Researcher at SAP and a master's candidate at the Hasso Plattner Institute (HPI), working on interpretability in multi-agent systems.

Prior to this, I graduated with a Bachelor of Technology (B.Tech) from SRM Institute of Science and Technology, Chennai (SRM IST), supervised by Prof. Dr. Athira M. Nambiar at the Center for AI in Computer Vision, where I focused on vision models, learning methodologies, and reasoning. I also received 'Outstanding Research Contribution Award 2025' [1 of 816] for research excellence.

Collaboration and discussion are welcomed.

Email  /  CV  /  LinkedIn /  Google Scholar  /  Github

profile photo
Research

My main research interests lie in computer vision and deep learning. I am particularly interested in general machine intelligence, learning methods for vision models and label-efficient (semi-supervised, unsupervised, self-supervised) approaches for deep learning for iterative self-improvement, interpretability and world modelling . I am always open to new ideas and collaborations in related fields. This website gives an overview of my recent research and background.

Highlights

05/2026

Paper on dual memory framework for agentic systems accepted at CVPR 2026 , arxiv (SOON). alt text

03/2026

Started working at SAP Labs, Potsdam as an Applied Researcher.. alt text

08/2025

Our Paper on Unsupervised Continual Learning got selected for Spotlight Presentation at ECAI 2025 .
Selected Publications

For the up-to-date publication list, please visit the Google Scholar page.

Replay to Remember (R2R): An Efficient Uncertainty-driven Unsupervised Continual Learning Framework Using Generative Replay
Sriram Mandalika , Jammigumpula Harsha Vardhan, Athira Nambiar
28th European Conference on Artificial Intelligence (ECAI), 2025
(Spotlight Paper)

Paper | Code | Website

PRIMEDrive-CoT: A Precognitive Chain-of-Thought Framework for Uncertainty-Aware Object Interaction in Driving Scene Scenario
Sriram Mandalika , Lalitha V and Athira Nambiar
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025

Paper | Code | Website

SegXAL: Explainable Active Learning for Semantic Segmentation in Driving Scene Scenarios
Sriram Mandalika and Athira Nambiar
27th International Conference on Pattern Recognition (ICPR), 2024

Paper | Code | Website

Academic Service

Reviewer: TMLR, ICCV (MCL Workshop 2025, XAI Workshop 2025), ICML 2025, ICML LatinX Workshop 2025, BMVC 2025.


Website template from here

* denotes equal contribution