Moritz Reuss — Robotics & VLA Research
Efficient generalist robot foundation models — Vision-Language-Action (VLA) models and World-Action Models (WAMs), diffusion/flow policies. Publications, projects, and writing.
Moritz Reuss I am a researcher at NVIDIA's Seattle Robotics Lab, based in Zürich, building efficient and versatile robot foundation models for manipulation. I recently completed my PhD with distinction (summa cum laude) at KIT's Intuitive Robots Lab, supervised by Rudolf Lioutikov. My goal is to make capable generalist robot foundation models efficient enough to train outside frontier labs and practical enough to run on local hardware. Email / CV / Google Scholar / GitHub / LinkedIn / Perspectives Full Publication List FLOWER: Democratizing Generalist Robot Policies with Efficient…
saved by
related reading
- State of Vision-Language-Action (VLA) Research at ICLR 2026 – Moritz Reussmbreuss.github.io
- Computer Vision and Geometry Group | Robot Learningcvg.ethz.ch
- What Matters for Latent Actions in Robot Learningcarldegio.github.io
- [2602.10098] VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Modelarxiv.org
- Pretrained to Imagine, Fine-Tuned to Act: The Rise of World-Action Models | NVIDIA Technical Blogdeveloper.nvidia.com
- Video models are zero-shot learners and reasonersarxiv.org
- Diffusion Policy: Visuomotor Policy Learning via Action Diffusiondiffusion-policy.cs.columbia.edu
- Emergence of Human to Robot Transfer in Vision-Language-Action Modelspi.website
- Causal Video Models Are Data-Efficient Robot Policy Learners | Rhoda AIrhoda.ai
- Precise Manipulation with Efficient Online RLpi.website
- Generalist - GEN-0 / Embodied Foundation Models That Scale with Physical Interactiongeneralistai.com
- A Unified Model for Motion-Conditioned Robot Co-designtransformer-transformer.github.io