Sven Elflein

I am a third-year PhD student at the University of Toronto, supervised by Sanja Fidler (Toronto AI Lab) and Laura Leal-Taixé(DVL).

Before that I've worked on machine learning models and uncertainty estimation to reason about future actions of other traffic participants for autonomous driving at Argo AI (RIP).

In 2021, I graduated with a Master's degree in Computer Science at Technical University of Munich, focusing on machine learning and mathematics. I wrote my thesis at DAML working with Prof. Günnemann.

profile photo

Research

My research interest lie in the areas of machine learning, uncertainty estimation, and generative models. Currently, I'm interested in an "understanding-through-synthesis" approach, leveraging generative models combined with interpretable priors, e.g., 3D geometry in computer vision and robotics, or physical constraints in drug discovery.

Addressable Memory for Video World Models
Xindi Wu, Sven Elflein, James Lucas, Olga Russakovsky, Laura Leal-Taixé, Despoina Paschalidou, Jonathan Lorraine, Aljoša Ošep
NeurIPS, 2026
Project page
FlowBender: Feedback-Aware Training for Self-Correcting Conditional Flows
Daniel Gilo, Sven Elflein, Ido Sobol, Or Litany
arXiv, 2026
Project page Code
Déjà View: Looping Transformers for Multi-View 3D Reconstruction
Alessandro Burzio, Tobias Fischer, Sven Elflein, Qunjie Zhou, Riccardo de Lutio, Jiawei Ren, Jiahui Huang, Shengyu Huang, Marc Pollefeys, Laura Leal-Taixé, Zan Gojcic, Haithem Turki
NeurIPS, 2026 — Oral (top 0.36% of submissions)
Project page Code
VGG-T3: Offline Feed-Forward 3D Reconstruction at Scale
Sven Elflein, Ruilong Li, Sérgio Agostinho, Zan Gojcic, Laura Leal-Taixé, Qunjie Zhou, Aljosa Osep
CVPR, 2026
Bibtex Project page Code
Effect of test-time inner-loop iterations on performance Test-Time Training with KV Binding Is Secretly Linear Attention
Junchen Liu, Sven Elflein, Or Litany, Zan Gojcic, Ruilong Li
ICML, 2026
Project page Code
LongPerceptualThoughts data synthesis pipeline LongPerceptualThoughts: Distilling System-2 Reasoning for System-1 Perception
Yuan-Hong Liao, Sven Elflein, Liu He, Laura Leal-Taixé, Yejin Choi, Sanja Fidler, David Acuna
COLM, 2025
Project page Code
Light3R-SfM: Towards Feed-forward Structure-from-Motion
Sven Elflein, Qunjie Zhou, Sérgio Agostinho, Laura Leal-Taixé
CVPR, 2025 — Spotlight (top 3% of submissions)
Project page
MATCHA: Towards Matching Anything
Fei Xue, Sven Elflein, Laura Leal-Taixé, Qunjie Zhou
CVPR (Highlight), 2025
Project page Code
MG-GAN: A Multi-Generator Model Preventing Out-of-Distribution Samples in Pedestrian Trajectory Prediction
Patrick Dendorfer*, Sven Elflein*, Laura Leal-Taixe
ICCV, 2021
Bibtex Project page Code
On Out-of-distribution Detection with Energy-based Models
Sven Elflein, Bertrand Charpentier, Daniel Zügner, Stephan Günnemann
ICML, UDM Workshop, 2021
Bibtex Project page Code