Sam Bond-Taylor

Research

I am a senior researcher in the Health Futures group at Microsoft Research Cambridge, researching multi-modal modals for medical imaging as part of the Biomedical Imaging Team. Some of the projects I have been involved in include RadEdit, an approach for stress-testing biomedical vision models by using diffusion models to edit images; and RAD-DINO), where we studied whether pre-training on multi-modal biomedical data can actually be a hindrance compared to uni-modal methods.

Prior to joining Microsoft I studied for my PhD at Durham University, where I explored how generative imaging models can be efficiently scaled to high resolutions, with papers including $\infty$-Diff and Unleashing Transformers.

Publications

image-left RadEdit: stress-testing biomedical vision models via diffusion image editing
European Conference on Computer Vision, 2024
F. Pérez-García, S. Bond-Taylor, P.P Sanchez, B. van Breugel, D.C. Castro, H. Sharma, V. Salvatelli, M.T. Wetscherek, H. Richardson, M.P. Lungren, A. Nori, J. Alvarez-Valle, O. Oktay, M. Ilse
[Model] [Paper]

image-left MAIRA-Seg: Enhancing Radiology Report Generation with Segmentation-Aware Multimodal Large Language Models
Machine Learning for Health (ML4H), 2024
H. Sharma, V. Salvatelli, S. Srivastav, K. Bouzid, S. Bannur, D.C. Castro, M. Ilse, S. Bond-Taylor, M. Prasanna Ranjit, F. Falck, F. Pérez-García, A. Schwaighofer, H. Richardson, M.T. Wetscherek, S.L. Hyland, J. Alvarez-Valle
[Paper]

image-left Maira-2: Grounded Radiology Report Generation
arXiv preprint 2406.04449, 2024
S. Bannur, K. Bouzid, D.C. Castro, A. Schwaighofer, A. Thieme, S. Bond-Taylor, M. Ilse, F. Pérez-García, V. Salvatelli, H. Sharma, F. Meissen, M. Ranjit, S. Srivastav, J. Gong, N.C.F. Codella, F. Falck, O. Oktay, M.P. Lungren, M.T. Wetscherek, J. Alvarez-Valle, and S.L. Hyland
[Model] [Paper]

image-left RAD-DINO: Exploring Scalable Medical Image Encoders Beyond Text Supervision
arXiv preprint arXiv:2401.10815, 2024
F. Pérez-García, H. Sharma, S. Bond-Taylor, K. Bouzid, V. Salvatelli, M. Ilse, S. Bannur, D.C. Castro, A. Schwaighofer, M.P. Lungren, M. Wetscherek, N. Codella, S.L. Hyland, J. Alvarez-Valle, and O. Oktay
[Model] [Paper]

$\infty$-Diff: Infinite Resolution Diffusion with Subsampled Mollified States
arXiv preprint arXiv:2303.18242, 2023
S. Bond-Taylor and C. G. Willcocks
[Code] [Paper]

Unaligned 2D to 3D Translation with Conditional Vector-Quantized Code Diffusion using Transformers
IEEE/CVF International Conference on Computer Vision, 2023
A. Corona-Figuera, S. Bond-Taylor, N. Bhowmik, Y. Gaus, T. P. Breckon, H. P. H. Shum and C. G. Willcocks

image-left Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codes
European Conference on Computer Vision, 2022
S. Bond-Taylor, P. Hessey, H. Sasaki, T. P. Breckon and C. G. Willcocks
[Project page] [Code] [Paper]

image-left MedNeRF: Medical Neural Radiance Fields for Reconstructing 3D-aware CT-Projections from a Single X-ray
IEEE Engineering in Medicine and Biology Conference, 2022
A. Corona-Figueroa, J. Frawley, S. Bond-Taylor, S. Bethapudi, H. P.H. Shum and C. G. Willcocks
[Code] [Paper]

image-left Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021
S. Bond-Taylor, A. Leach, Y. Long and C. G. Willcocks
[Paper]

image-left Gradient Origin Networks
International Conference on Learning Representations, 2021
S. Bond-Taylor and C. G. Willcocks
[Project page] [Code] [Paper]

image-left Shape Tracing: An extension of sphere tracing for 3D non-convex collision in protein docking
IEEE International Conference on Bioinformatics and Bioengineering 2020
A. Leach, L. S.P. Rudden, S. Bond-Taylor, J. C. Brigham, M. T. Degiacomi and C. G. Willcocks [Paper]