Scholar
Christian Schroeder de Witt
Google Scholar ID: DE60h_0AAAAJ
University of Oxford
Multi-agent Learning
Security
Safety
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
7,448
H-index
20
i10-index
34
Publications
20
Co-authors
12
list available
Contact
Email
clinic@wittlab.ai
GitHub
Open ↗
Publications
32 items
Weight Oracles: Reading Neural Network Weights with Language Models
2026
Cited
0
Can CaMeLs Talk? Securing Multi-Agent Systems Against Indirect Prompt Injection Attacks
2026
Cited
0
Reflections and Fragments: Securing LLMs Against Sequential Mosaic Attacks
2026
Cited
0
MA-JEPA: Joint-Embedding World Models for Multi-Agent Reinforcement Learning
2026
Cited
0
ORBIT: A Framework for Multi-Agent Safety and Security Evaluations
2026
Cited
0
Epsilon-Nash Equilibria in History-Dependent SA-MDPs
2026
Cited
0
Entangled Representations Amplify Collateral Damage in Unlearning
2026
Cited
0
Multimodal Model Diffing for Feature Discovery and Control
2026
Cited
0
Load more
Co-authors
10 total
Jakob Foerster
Associate Professor, University of Oxford
Philip Torr
Professor, University of Oxford
Shimon Whiteson
Professor of Computer Science, University of Oxford / Senior Staff Research Scientist, Waymo
Mikayel Samvelyan
Google DeepMind
Gregory Farquhar
DeepMind
Tabish Rashid
Microsoft Research
Wendelin Böhmer
Sequential Decision Making Group, Delft University of Technology
Bei Peng
Lecturer (Assistant Professor), University of Sheffield
