BB
Boris van Breugel
cs.LGcs.AIcs.CVstat.MLcs.CLcs.CRcs.CYcs.NEquant-ph
On Valency
published · living versionsW_r72x54u6·v1 · currentpublished
A Unified Hessian-Guided Framework for Post-Training Quantization
with Mads Toftrup, Marco Federici, Paul Whatmough, Markus Nagel
1 version
W_5txyfx4z·v1published
Preprints & journals
21 papers in the corpus · 2021–2026MobileWan: Closing the Quality Gap for Mobile Video Diffusion2607.06173v2 · Mohsen Ghafoorian, Denis Korzhenkov, Adil Karjauv et al.2026 · 0 citationsarXiv
Dissecting Quantization Error: A Concentration-Alignment Perspective2603.04359v2 · Marco Federici, Boris van Breugel, Paul Whatmough et al.2026 · 0 citationsarXiv
STaMP: Sequence Transformation and Mixed Precision for Low-Precision Activation Quantization2510.26771v1 · Marco Federici, Riccardo Del Chiaro, Boris van Breugel et al.2025 · 0 citationsarXiv
HadaNorm: Diffusion Transformer Quantization through Mean-Centered Transformations2506.09932v2 · Marco Federici, Riccardo Del Chiaro, Boris van Breugel et al.2025 · 0 citationsarXiv
FPTQuant: Function-Preserving Transforms for LLM Quantization2506.04985v2 · Boris van Breugel, Yelysei Bondarenko, Paul Whatmough et al.2025 · 0 citationsarXiv
Position: All Current Generative Fidelity and Diversity Metrics are Flawed2505.22450v1 · Ossi Raisa, Boris van Breugel, Mihaela van der Schaar2025 · 0 citationsarXiv
LaTable: Towards Large Tabular Models2406.17673v1 · Boris van Breugel, Jonathan Crabb'e, Rob Davis et al.2024 · 0 citationsarXiv
Why Tabular Foundation Models Should Be a Research Priority2405.01147v2 · Boris van Breugel, Mihaela van der Schaar2024 · 5 citationsarXiv
RadEdit: stress-testing biomedical vision models via diffusion image editing2312.12865v3 · Fernando P'erez-Garc'ia, Sam Bond-Taylor, Pedro P. Sanchez et al.2023 · 16 citationsEuropean Conference on Computer Vision (ECCV) 2024
Curated LLM: Synergy of LLMs and Data Curation for tabular augmentation in low-data regimes2312.12112v3 · Nabeel Seedat, Nicolas Huynh, Boris van Breugel et al.2023 · 1 citationarXiv
Can You Rely on Your Model Evaluation? Improving Model Evaluation with Synthetic Test Data2310.16524v1 · Boris van Breugel, Nabeel Seedat, Fergus Imrie et al.2023 · 6 citationsarXiv
Soft Mixture Denoising: Beyond the Expressive Bottleneck of Diffusion Models2309.14068v3 · Yangming Li, Boris van Breugel, Mihaela van der Schaar2023 · 0 citationsarXiv
The Spherical Grasshopper Problem2307.05359v1 · Boris van Breugel2023 · 0 citationsarXiv
Synthetic data, real errors: how (not) to publish and use synthetic data2305.09235v2 · Boris van Breugel, Zhaozhi Qian, Mihaela van der Schaar2023 · 15 citationsarXiv
Beyond Privacy: Navigating the Opportunities and Challenges of Synthetic Data2304.03722v1 · Boris van Breugel, Mihaela van der Schaar2023 · 14 citationsarXiv
Membership Inference Attacks against Synthetic Data through Overfitting Detection2302.12580v1 · Boris van Breugel, Hao Sun, Zhaozhi Qian et al.2023 · 23 citationsarXiv
Practical Approaches for Fair Learning with Multitype and Multivariate Sensitive Attributes2211.06138v1 · Tennison Liu, Alex J. Chan, Boris van Breugel et al.2022 · 0 citationsarXiv
What is Flagged in Uncertainty Quantification? Latent Density Models for Uncertainty Categorization2207.05161v2 · Hao Sun, Boris van Breugel, Jonathan Crabbe et al.2022 · 1 citationarXiv
DECAF: Generating Fair Synthetic Data Using Causally-Aware Generative Networks2110.12884v2 · Boris van Breugel, Trent Kyono, Jeroen Berrevoets et al.2021 · 30 citationsarXiv
How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models2102.08921v2 · Ahmed M. Alaa, Boris van Breugel, Evgeny Saveliev et al.2021 · 75 citationsarXiv
Stereotype and Skew: Quantifying Gender Bias in Pre-trained and Fine-tuned Language Models2101.09688v2 · Daniel de Vassimon Manela, David Errington, Thomas Fisher et al.2021 · 59 citationsarXiv
Career total: 25 works. 21 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.