1 citations · 3 across the 16 of their papers we have counts for
23 papers
KV-Kaizen: Learning Context-Adaptive Cache Compression Choices
Joao Monteiro, Louis Béthune, Anastasiia Filippova +3
As the context size of text processed with an LLM grows, the size of KV caches can outstrip the memory allocated for the original model weights. This impacts LLM throughput negativ…
Cartridges++: KV Cache Compression without Off-Context Derailment
Sonia Laguna, Joao Monteiro, Marco Cuturi +2
Serving long documents to a Large Language Model (LLM) repeatedly is expensive: computations grow with context length, and the memory footprint of the key-value (KV) cache balloons…
ConRAG: Lightweight inference of multi-hop relations
Kilian Bänziger, Sonia Laguna, Markus Kreft +4
Understanding how two entities are connected often requires tracing multi-hop relations across documents to identify intermediate entities and supporting evidence that explain a co…
Cyclostationary Phase Conditioning for Medical Time Series Diffusion
Samuel Ruiperez-Campillo, Michele Copetti, Jorge da Silva Goncalves +3
Many physiological time series, such as cardiac and brain recordings, exhibit cyclostationarity: their statistics vary periodically with an underlying cycle phase. Corruption from…
DynaMiCS: Fine-tuning LLMs with Performance Constraints using Dynamic Mixtures
Eleonora Gualdoni, Sonia Laguna, Louis Bethune +3
Multi-domain fine-tuning of large language models requires improving performance on target domains while preserving previous capabilities, such as general knowledge, instruction fo…
Foundation Model for Cardiac Time Series via Masked Latent Attention
Moritz Vandenhirtz, Samuel Ruipérez-Campillo, Simon Böhi +6
Electrocardiograms (ECGs) are among the most widely available clinical signals and play a central role in cardiovascular diagnosis. While recent foundation models (FMs) have shown…
Beyond Independent Frames: Latent Attention Masked Autoencoders for Multi-View Echocardiography
Simon Böhi, Irene Cannistraci, Sergio Muñoz Gonzalez +8
Echocardiography is a widely used modality for cardiac assessment due to its non-invasive and cost-effective nature, but the sparse and heterogeneous spatiotemporal views of the he…
Models Designed to Forget: Machine Unlearning via Key Deletion
Sonia Laguna, Jorge da Silva Goncalves, Moritz Vandenhirtz +3
Machine unlearning for vision models is rapidly becoming a practical requirement, driven by privacy regulations, data errors, and the need to remove harmful or corrupted training i…
Steering Generative Models for Accessibility: EasyRead Image Generation
Nicolas Dickenmann, Yanis Merzouki, Sonia Laguna +4
EasyRead pictograms are simple, visually clear images that represent specific concepts and support comprehension for people with intellectual disabilities, low literacy, or languag…
Reference-Guided Machine Unlearning
Jonas Mirlach, Sonia Laguna, Julia E. Vogt
Machine unlearning aims to remove the influence of specific training data from a model while preserving its general utility. In vision, many approximate unlearning methods pursue t…
Structure is Supervision: Multiview Masked Autoencoders for Radiology
Sonia Laguna, Andrea Agostini, Alain Ryser +9
Building robust medical machine learning systems requires pretraining strategies that exploit the intrinsic structure present in clinical data. We introduce Multiview Masked Autoen…
Post-hoc Stochastic Concept Bottleneck Models
Wiktor Jan Hoffmann, Sonia Laguna, Moritz Vandenhirtz +2
Concept Bottleneck Models (CBMs) are interpretable models that predict the target variable through high-level human-understandable concepts, allowing users to intervene on mispredi…
Leveraging the Structure of Medical Data for Improved Representation Learning
Andrea Agostini, Sonia Laguna, Alain Ryser +7
Building generalizable medical AI systems requires pretraining strategies that are data-efficient and domain-aware. Unlike internet-scale corpora, clinical datasets such as MIMIC-C…
Interpretable Reward Modeling with Active Concept Bottlenecks
Sonia Laguna, Katarzyna Kobalczyk, Julia E. Vogt +1
We introduce Concept Bottleneck Reward Models (CB-RM), a reward modeling framework that enables interpretable preference learning through selective concept annotation. Unlike stand…
Uncertainty Estimation for Trust Attribution to Speed-of-Sound Reconstruction with Variational Networks
Sonia Laguna, Lin Zhang, Can Deniz Bezek +4
Speed-of-sound (SoS) is a biomechanical characteristic of tissue, and its imaging can provide a promising biomarker for diagnosis. Reconstructing SoS images from ultrasound acquisi…
Text To 3D Object Generation For Scalable Room Assembly
Sonia Laguna, Alberto Garcia-Garcia, Marie-Julie Rakotosaona +3
Modern machine learning models for scene understanding, such as depth estimation and object tracking, rely on large, high-quality datasets that mimic real-world deployment scenario…
Beyond Glucose-Only Assessment: Advancing Nocturnal Hypoglycemia Prediction in Children with Type 1 Diabetes
Marco Voegeli, Sonia Laguna, Heike Leutheuser +3
The dead-in-bed syndrome describes the sudden and unexplained death of young individuals with Type 1 Diabetes (T1D) without prior long-term complications. One leading hypothesis at…
Measuring Leakage in Concept-Based Methods: An Information Theoretic Approach
Mikael Makonnen, Moritz Vandenhirtz, Sonia Laguna +1
Concept Bottleneck Models (CBMs) aim to enhance interpretability by structuring predictions around human-understandable concepts. However, unintended information leakage, where pre…
RadVLM: A Multitask Conversational Vision-Language Model for Radiology
Nicolas Deperrois, Hidetoshi Matsuo, Samuel Ruipérez-Campillo +12
The widespread use of chest X-rays (CXRs), coupled with a shortage of radiologists, has driven growing interest in automated CXR analysis and AI-assisted reporting. While existing…
Exploiting Interpretable Capabilities with Concept-Enhanced Diffusion and Prototype Networks
Alba Carballo-Castro, Sonia Laguna, Moritz Vandenhirtz +1
Concept-based machine learning methods have increasingly gained importance due to the growing interest in making neural networks interpretable. However, concept annotations are gen…
Towards Scalable Newborn Screening: Automated General Movement Assessment in Uncontrolled Settings
Daphné Chopard, Sonia Laguna, Kieran Chin-Cheong +4
General movements (GMs) are spontaneous, coordinated body movements in infants that offer valuable insights into the developing nervous system. Assessed through the Prechtl GM Asse…
Stochastic Concept Bottleneck Models
Moritz Vandenhirtz, Sonia Laguna, Ričards Marcinkevičs +1
Concept Bottleneck Models (CBMs) have emerged as a promising interpretable method whose final prediction is based on intermediate, human-understandable concepts rather than the raw…
Beyond Concept Bottleneck Models: How to Make Black Boxes Intervenable?
Sonia Laguna, Ričards Marcinkevičs, Moritz Vandenhirtz +1
Recently, interpretable machine learning has re-explored concept bottleneck models (CBM). An advantage of this model class is the user's ability to intervene on predicted concept v…