arXiv subjects

Benoit Schmauch

Publications and source records attributed to Benoit Schmauch.

6 recordsLinked to original sources

arXiv2026

Robustifying pathology foundation models via fine-tuning

Pathology foundation models (FMs) produce powerful tile-level representations which remain sensitive to scanner and staining variability, undermining deployment across laboratories. We develop a novel fine-tuning recipe that improves the robustness of pathology FMs to acquisition factors. Applied to ten different FMs, our fine-tuning strategy consistently improves robustness for every model as well as downstream performance, with no observed trade-off. On average, it raises the PathoROB robustness index by 23% (from 0.72 to 0.87) and increases the overall cross-benchmark performance by 43% on Patho-Bench, HEST and THUNDER combined, with individual gains reaching up to 72% in robustness (Phikon-v2) and 76% in performance (Midnight-12k). We publicly release the fine-tuned versions of Phikon-v2 (Phaet) and Midnight-12k (Mascaret) at https://huggingface.co/wearewaiv/models.

cs.CV↗

arXiv2024

CARMIL: Context-Aware Regularization on Multiple Instance Learning models for Whole Slide Images

Multiple Instance Learning (MIL) models have proven effective for cancer prognosis from Whole Slide Images. However, the original MIL formulation incorrectly assumes the patches of the same image to be independent, leading to a loss of spatial context as information flows through the network. Incorporating contextual knowledge into predictions is particularly important given the inclination for cancerous cells to form clusters and the presence of spatial indicators for tumors. State-of-the-art methods often use attention mechanisms eventually combined with graphs to capture spatial knowledge. In this paper, we take a novel and transversal approach, addressing this issue through the lens of regularization. We propose Context-Aware Regularization for Multiple Instance Learning (CARMIL), a versatile regularization scheme designed to seamlessly integrate spatial knowledge into any MIL model. Additionally, we present a new and generic metric to quantify the Context-Awareness of any MIL model when applied to Whole Slide Images, resolving a previously unexplored gap in the field. The efficacy of our framework is evaluated for two survival analysis tasks on glioblastoma (TCGA GBM) and colon cancer data (TCGA COAD).

cs.CV↗

arXiv2021

Self supervised learning improves dMMR/MSI detection from histology slides across multiple cancers

Microsatellite instability (MSI) is a tumor phenotype whose diagnosis largely impacts patient care in colorectal cancers (CRC), and is associated with response to immunotherapy in all solid tumors. Deep learning models detecting MSI tumors directly from H&E stained slides have shown promise in improving diagnosis of MSI patients. Prior deep learning models for MSI detection have relied on neural networks pretrained on ImageNet dataset, which does not contain any medical image. In this study, we leverage recent advances in self-supervised learning by training neural networks on histology images from the TCGA dataset using MoCo V2. We show that these networks consistently outperform their counterparts pretrained using ImageNet and obtain state-of-the-art results for MSI detection with AUCs of 0.92 and 0.83 for CRC and gastric tumors, respectively. These models generalize well on an external CRC cohort (0.97 AUC on PAIP) and improve transfer from one organ to another. Finally we show that predictive image regions exhibit meaningful histological patterns, and that the use of MoCo features highlighted more relevant patterns according to an expert pathologist.

eess.IV↗

arXiv2018

CNN+LSTM Architecture for Speech Emotion Recognition with Data Augmentation

In this work we design a neural network for recognizing emotions in speech, using the IEMOCAP dataset. Following the latest advances in audio analysis, we use an architecture involving both convolutional layers, for extracting high-level features from raw spectrograms, and recurrent ones for aggregating long-term dependencies. We examine the techniques of data augmentation with vocal track length perturbation, layer-wise optimizer adjustment, batch normalization of recurrent layers and obtain highly competitive results of 64.5% for weighted accuracy and 61.7% for unweighted accuracy on four emotions.

cs.SD↗

arXiv2013

The Physics of Neutrinos

These lecture notes are based on a course given at Institut de Physique Théorique of CEA/Saclay in January/February 2013.

hep-ph↗

arXiv2011

Linear Growth of Structure in the Symmetron Model

In the symmetron mechanism, the fifth force mediated by a coupled scalar field (the symmetron) is suppressed in high-density regions due to the restoration of symmetry in the symmetron potential. In this paper we study the background cosmology and large scale structure formation in the linear perturbation regime of the symmetron model. Analytic solutions to the symmetron in the cosmological background are found, which agree well with numerical results. We discuss the effect of the symmetron perturbation on the growth of matter perturbation, in particular the implications of the brief period of tachyonic instability caused by the negative mass squared of the symmetron during symmetry breaking. Our analysis and numerical results show that this instability has only very small effects on the growth of structures on sub-horizon scales, and even at horizon scales its influence is not as drastic as naively expected. The symmetron fifth force in the non-tachyonic regime does affect the formation of structure in a nontrivial way which could be cosmologically observable.

astro-ph.CO↗