Constantin Venhoff
Scholar

Constantin Venhoff

Google Scholar ID: kzHpG-EAAAAJ
University of Oxford
AIMLInterpretabilityMechanistic InterpretabilityAI Alignment
Google Scholar↗
Citations & Impact
All-time
Citations
28
 
H-index
2
 
i10-index
1
 
Publications
8
 
Co-authors
9
list available
Contact
No contact links provided.
Publications
9 items
Towards Understanding Multimodal Fine-Tuning: Spatial Features
2026
Cited
0
Sparse CLIP: Co-Optimizing Interpretability and Performance in Contrastive Learning
2026
Cited
0
Too Late to Recall: Explaining the Two-Hop Problem in Multimodal Knowledge Retrieval
2025
Cited
0
Base Models Know How to Reason, Thinking Models Learn When
2025
Cited
0
Towards Mechanistic Defenses Against Typographic Attacks in CLIP
2025
Cited
0
Reasoning-Finetuning Repurposes Latent Representations in Base Models
2025
Cited
0
Understanding Reasoning in Thinking Language Models via Steering Vectors
2025
Cited
0
How Visual Representations Map to Language Feature Space in Multimodal LLMs
2025
Cited
0
Resume (English only)
Co-authors
9 total
Philip Torr
Philip Torr
Professor, University of Oxford
Neel Nanda
Neel Nanda
Mechanistic Interpretability Team Lead, Google DeepMind
Ashkan Khakzar
Ashkan Khakzar
University of Oxford
Co-author 4
Co-author 4
Bernhard Rumpe
Bernhard Rumpe
RWTH Aachen University
Christian Schroeder de Witt
Christian Schroeder de Witt
University of Oxford
Iván Arcuschin
Iván Arcuschin
Independent Researcher
Arthur Conmy
Arthur Conmy
Google DeepMind