Lokesh Veeramacheneni

I am a PhD student at Universität Bonn, advised by Prof. Dr. Juergen Gall and currently employed by HPC/A-Lab. My research centers on computer vision, with focus on Diffusion models and efficient fine-tuning of vision models. I collaborate with researchers including Prof. Dr. Hilde Kuehne and Dr. Moritz Wolter.

Previously, I worked as a Software Developer at RE: GmbH with focus on Ruby on Rails and as a SAP ABAP developer at Infosys Ltd. I hold Master of Science degree in Autonomous Systems from Hochschule Bonn-Rhein-Sieg, Germany and Bachelors in Electronics and Communications Engineering from KL University, India.

Email / Scholar / Github

Publications

	Canonical Rank Adaptation (CaRA): An Efficient Fine-Tuning Strategy for Vision Transformers Lokesh Veeramacheneni, Moritz Wolter, Hilde Kuehne, Juergen Gall Forty-Second International Conference on Machine Learning, ICML 2025 Project Page / Paper / Code
	More Rigorous Software Engineering Would Improve Reproducibility in Machine Learning Research Moritz Wolter, Lokesh Veeramacheneni, Charles Tapley Hoyt arXiv, 2025 Paper / Code
	Fréchet Wavelet Distance (FWD): A Domain-Agnostic Metric for Image Generation Lokesh Veeramacheneni, Moritz Wolter, Hilde Kuehne, Juergen Gall Thirteenth International Conference on Learning Representations, ICLR 2025 Project Page / Paper / Code / Pip
	On the Stability of Neural Segmentation in Radiology Moritz Wolter, Lokesh Veeramacheneni, Bettina Baeßler, Ulrike I Attenberger, Barbara D Wichtmann European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2024 Paper / Code
	Canonical Convolutional Neural Networks Lokesh Veeramacheneni, Moritz Wolter, Reinhard Klein, Jochen Garcke International Joint Conference on Neural Networks, IJCNN 2022 Paper / Code
	Fabrication of highly sensitive and selective nanocomposite film based on CuNPs/fullerene-C₆₀/MWCNTs: An electrochemical nanosensor for trace recognition of paracetamol Pradeep Kumar Brahman, Lakkavarapu Suresh, Lokesh Veeramacheneni, Syed Nizamuddin Analytica Chemica Acta, Volume 917, 2016, Pages 107-116 Paper

Workshops

	Enhancing Explainability with Multimodal Context Representations for Smarter Robots Anargh Viswanath, Lokesh Veeramacheneni, Hendrik Buschmeier 3rd Workshop on Explainability in HRC, 2025 Paper * denotes equal contribution
	A Benchmark for Out of Distribution Detection in Point Cloud 3D Semantic Segmentation Lokesh Veeramacheneni, Matias Valdenegro Toro NeurIPS Workshop on Robot Learning: Trustworthy Robotics, 2022 Paper

Supervised Theses

— Parameter-Efficient Adaptation of Open-Source Language Models for Clinical MRI Protocol Automation, Masterarbeit Informatik, Zahra Ganji, 2025

Academic Service

Reviewer

CVPR 2026, ECCV 2026, NeurIPS 2026, TMLR 2026

Teaching

University of Bonn	Part of tutoring team for - Foundations of Machine Learning SoSe23, WiSe23, WiSe24, WiSe25 - Foundations of Machine Learning for Principal Investigators WiSe23, SoSe25 - Advanced Machine Learning SoSe24, SoSe25 - C++ and CUDA Programming for Machine Learning SoSe25
Hochschule Bonn-Rhein-Sieg	C++ section in Foundation course WiSe19, SoSe20, WiSe20, SoSe21

Website design adopted from Jon Barron.