Benjamin Schneider

I’m a first-year CS master student masters student at University of Waterloo, advised by Wenhu Chen (TIGER-Lab) and Florian Kerschbaum.

I’m broadly interested in foundation models that understand the world (however I chose to define that). My specific interests are in unified methods for multimodal representation learning across many modalities (audio/video/text/images) and (increasingly) small-scale world models. Sometimes I do work in systems for multimodal ML, mostly in an effort to squeeze as much as possible out of limited hardware.

I’m pretty terrible about keeping my website updated. 😅
So, for an up-to-date list of publications please check my scholar, my code/projects are hosted under the TIGER-Lab github.

News

May 15, 2025	First public release of QuickVideo, our library for efficient (long) VideoLLM inference. QuickVideo is an ongoing project focused on improving systems and models for VideoLLMs, please provide feedback if there are features you want implemented!
Mar 04, 2025	We release ABC, a model text-guided visual retrieval.

Publications

Universal Backdoor Attacks

Benjamin Schneider, Nils Lukas, and Florian Kerschbaum

In The Twelfth International Conference on Learning Representations, 2024

arXiv Bib Code

@inproceedings{schneider2024universal,
  title = {Universal Backdoor Attacks},
  author = {Schneider, Benjamin and Lukas, Nils and Kerschbaum, Florian},
  booktitle = {The Twelfth International Conference on Learning Representations},
  year = {2024},
  url = {https://openreview.net/forum?id=3QkzYBSWqL},
}