arXiv subjects

Marc Pouly

Publications and source records attributed to Marc Pouly.

At least 19 recordsLinked to original sources

arXiv2026

Picking the Right Image to Classify: Reliable-Input Selection in Teledermatology

Dermatology models face distribution shifts in teledermatology settings, where submitted images differ from the training data in lighting, angle, distance, focus, and framing. These test-time images are ordinary clinical photographs, but some fall outside the model's training conditions, leading the model to often misclassify them due to shifts in acquisition between training and deployment. When multiple images of the same case exist (several photos of one patient or lesion), a natural way to improve accuracy is therefore to select the image the model is most likely to classify correctly. We call this task reliable-input selection. An oracle that, for each case, selects a correctly classified image when one exists raises weighted F1 by about 20 percentage points on average across six dermatology datasets and nine frozen backbones. This oracle is an upper bound that sees the labels, whereas a selector must choose blindly. Capturing this gain in practice is hard. A selector that needs no pretraining data applies to any frozen model, including those whose data is not public. It must judge reliability from quantities the model exposes at inference: its embeddings, their norms, and its confidence. We benchmark four such training-data-free selectors: the embedding norm, the neighborhood consensus among a case's images, the stability of the prediction under small perturbations, and the model's own confidence. No training-data-free selector substantially narrows this oracle gap. The best of them is the model's own confidence, but it recovers only a small part of the gap on the clinical datasets. A small labeled reference set does not help either: the best selector overall, a fusion of confidence and Mahalanobis distance, still leaves most of the gap. To our knowledge, this is the first study to introduce and benchmark reliable input selection, a clinically important, unsolved task.

cs.CV↗

arXiv2026

Consensus Measures for Unstructured Biomedical Text Annotations

Biomedical literature is increasingly mined for knowledge beyond the questions it was written to answer. Because the target concepts are not known in advance, annotators prefer open-ended labels, whose agreement is hard to quantify. We study soft inter-rater reliability for annotators providing unstructured texts for biomedical annotation tasks. Synthetic experiments show that soft reliability can be quantified using a variety of semantic equivalence measures, and that the choice of measure affects failure modes of the estimation. Embeddings are scalable, but limited when differentiating similar but distinct concepts. Large language models are promising, but limited by scalability for estimating agreement by chance. Finally, we suggest measures based on natural language inference as a sensible compromise.

cs.CL↗

arXiv2025

A Global Atlas of Digital Dermatology to Map Innovation and Disparities

The adoption of artificial intelligence in dermatology promises democratized access to healthcare, but model reliability depends on the quality and comprehensiveness of the data fueling these models. Despite rapid growth in publicly available dermatology images, the field lacks quantitative key performance indicators to measure whether new datasets expand clinical coverage or merely replicate what is already known. Here we present SkinMap, a multi-modal framework for the first comprehensive audit of the field's entire data basis. We unify the publicly available dermatology datasets into a single, queryable semantic atlas comprising more than 1.1 million images of skin conditions and quantify (i) informational novelty over time, (ii) dataset redundancy, and (iii) representation gaps across demographics and diagnoses. Despite exponential growth in dataset sizes, informational novelty across time has somewhat plateaued: Some clusters, such as common neoplasms on fair skin, are densely populated, while underrepresented skin types and many rare diseases remain unaddressed. We further identify structural gaps in coverage: Darker skin tones (Fitzpatrick V-VI) constitute only 5.8% of images and pediatric patients only 3.0%, while many rare diseases and phenotype combinations remain sparsely represented. SkinMap provides infrastructure to measure blind spots and steer strategic data acquisition toward undercovered regions of clinical space.

cs.DL↗

arXiv2025

Representation-Based Data Quality Audits for Audio

Data quality issues such as off-topic samples, near duplicates, and label errors often limit the performance of audio-based systems. This paper addresses these issues by adapting SelfClean, a representation-to-rank data auditing framework, from the image to the audio domain. This approach leverages self-supervised audio representations to identify common data quality issues, creating ranked review lists that surface distinct issues within a single, unified process. The method is benchmarked on the ESC-50, GTZAN, and a proprietary industrial dataset, using both synthetic and naturally occurring corruptions. The results demonstrate that this framework achieves state-of-the-art ranking performance, often outperforming issue-specific baselines and enabling significant annotation savings by efficiently guiding human review.

cs.SD↗

arXiv2025

Clinical Uncertainty Impacts Machine Learning Evaluations

Clinical dataset labels are rarely certain as annotators disagree and confidence is not uniform across cases. Typical aggregation procedures, such as majority voting, obscure this variability. In simple experiments on medical imaging benchmarks, accounting for the confidence in binary labels significantly impacts model rankings. We therefore argue that machine-learning evaluations should explicitly account for annotation uncertainty using probabilistic metrics that directly operate on distributions. These metrics can be applied independently of the annotations' generating process, whether modeled by simple counting, subjective confidence ratings, or probabilistic response models. They are also computationally lightweight, as closed-form expressions have linear-time implementations once examples are sorted by model score. We thus urge the community to release raw annotations for datasets and to adopt uncertainty-aware evaluation so that performance estimates may better reflect clinical data.

cs.AI↗

arXiv2025

Is Hyperbolic Space All You Need for Medical Anomaly Detection?

Medical anomaly detection has emerged as a promising solution to challenges in data availability and labeling constraints. Traditional methods extract features from different layers of pre-trained networks in Euclidean space; however, Euclidean representations fail to effectively capture the hierarchical relationships within these features, leading to suboptimal anomaly detection performance. We propose a novel yet simple approach that projects feature representations into hyperbolic space, aggregates them based on confidence levels, and classifies samples as healthy or anomalous. Our experiments demonstrate that hyperbolic space consistently outperforms Euclidean-based frameworks, achieving higher AUROC scores at both image and pixel levels across multiple medical benchmark datasets. Additionally, we show that hyperbolic space exhibits resilience to parameter variations and excels in few-shot scenarios, where healthy images are scarce. These findings underscore the potential of hyperbolic space as a powerful alternative for medical anomaly detection. The project website can be found at https://hyperbolic-anomalies.github.io

cs.CV↗

arXiv2025

FoodTaxo: Generating Food Taxonomies with Large Language Models

We investigate the utility of Large Language Models for automated taxonomy generation and completion specifically applied to taxonomies from the food technology industry. We explore the extent to which taxonomies can be completed from a seed taxonomy or generated without a seed from a set of known concepts, in an iterative fashion using recent prompting techniques. Experiments on five taxonomies using an open-source LLM (Llama-3), while promising, point to the difficulty of correctly placing inner nodes.

cs.CL↗

arXiv2025

CleanPatrick: A Benchmark for Image Data Cleaning

Robust machine learning depends on clean data, yet current image data cleaning benchmarks rely on synthetic noise or narrow human studies, limiting comparison and real-world relevance. We introduce CleanPatrick, the first large-scale benchmark for data cleaning in the image domain, built upon the publicly available Fitzpatrick17k dermatology dataset. We collect 496,377 binary annotations from 933 medical crowd workers, identify off-topic samples (4%), near-duplicates (21%), and label errors (32%), and employ an aggregation model inspired by item-response theory followed by expert review to derive high-quality ground truth. CleanPatrick formalizes issue detection as a ranking task and employs standard ranking metrics that mirror real audit workflows. We benchmark classical anomaly detectors, perceptual hashing, SSIM, Confident Learning, NoiseRank, FINE, BHN, and SelfClean. On CleanPatrick, self-supervised representations excel at near-duplicate detection, classical methods achieve competitive off-topic detection under constrained review budgets, and detecting implausible labels under conservative human judgment remains challenging for fine-grained medical classification. By releasing both the dataset and the evaluation framework, CleanPatrick enables a systematic comparison of image-cleaning strategies.

cs.CV↗

arXiv2025

Reference-Free Evaluation of Taxonomies

We introduce two reference-free metrics for quality evaluation of taxonomies in the absence of labels. The first metric evaluates robustness by calculating the correlation between semantic and taxonomic similarity, addressing error types not considered by existing metrics. The second uses Natural Language Inference to assess logical adequacy. Both metrics are tested on five taxonomies and are shown to correlate well with F1 against ground truth taxonomies. We further demonstrate that our metrics can predict downstream performance in hierarchical classification when used with label hierarchies.

cs.CL↗

arXiv2024

Towards Scalable Foundation Models for Digital Dermatology

The growing demand for accurate and equitable AI models in digital dermatology faces a significant challenge: the lack of diverse, high-quality labeled data. In this work, we investigate the potential of domain-specific foundation models for dermatology in addressing this challenge. We utilize self-supervised learning (SSL) techniques to pre-train models on a dataset of over 240,000 dermatological images from public and private collections. Our study considers several SSL methods and compares the resulting foundation models against domain-agnostic models like those pre-trained on ImageNet and state-of-the-art models such as MONET across 12 downstream tasks. Unlike previous research, we emphasize the development of smaller models that are more suitable for resource-limited clinical settings, facilitating easier adaptation to a broad range of use cases. Results show that models pre-trained in this work not only outperform general-purpose models but also approach the performance of models 50 times larger on clinically relevant diagnostic tasks. To promote further research in this direction, we publicly release both the training code and the foundation models, which can benefit clinicians in dermatological applications.

cs.CV↗

arXiv2024

PASSION for Dermatology: Bridging the Diversity Gap with Pigmented Skin Images from Sub-Saharan Africa

Africa faces a huge shortage of dermatologists, with less than one per million people. This is in stark contrast to the high demand for dermatologic care, with 80% of the paediatric population suffering from largely untreated skin conditions. The integration of AI into healthcare sparks significant hope for treatment accessibility, especially through the development of AI-supported teledermatology. Current AI models are predominantly trained on white-skinned patients and do not generalize well enough to pigmented patients. The PASSION project aims to address this issue by collecting images of skin diseases in Sub-Saharan countries with the aim of open-sourcing this data. This dataset is the first of its kind, consisting of 1,653 patients for a total of 4,901 images. The images are representative of telemedicine settings and encompass the most common paediatric conditions: eczema, fungals, scabies, and impetigo. We also provide a baseline machine learning model trained on the dataset and a detailed performance analysis for the subpopulations represented in the dataset. The project website can be found at https://passionderm.github.io/.

cs.CV↗

arXiv2024

Hyperbolic Metric Learning for Visual Outlier Detection

Out-Of-Distribution (OOD) detection is critical to deploy deep learning models in safety-critical applications. However, the inherent hierarchical concept structure of visual data, which is instrumental to OOD detection, is often poorly captured by conventional methods based on Euclidean geometry. This work proposes a metric framework that leverages the strengths of Hyperbolic geometry for OOD detection. Inspired by previous works that refine the decision boundary for OOD data with synthetic outliers, we extend this method to Hyperbolic space. Interestingly, we find that synthetic outliers do not benefit OOD detection in Hyperbolic space as they do in Euclidean space. Furthermore we explore the relationship between OOD detection performance and Hyperbolic embedding dimension, addressing practical concerns in resource-constrained environments. Extensive experiments show that our framework improves the FPR95 for OOD detection from 22\% to 15\% and from 49% to 28% on CIFAR-10 and CIFAR-100 respectively compared to Euclidean methods.

cs.CV↗

arXiv2024

Estimating Text Similarity based on Semantic Concept Embeddings

Due to their ease of use and high accuracy, Word2Vec (W2V) word embeddings enjoy great success in the semantic representation of words, sentences, and whole documents as well as for semantic similarity estimation. However, they have the shortcoming that they are directly extracted from a surface representation, which does not adequately represent human thought processes and also performs poorly for highly ambiguous words. Therefore, we propose Semantic Concept Embeddings (CE) based on the MultiNet Semantic Network (SN) formalism, which addresses both shortcomings. The evaluation on a marketing target group distribution task showed that the accuracy of predicted target groups can be increased by combining traditional word embeddings with semantic CEs.

cs.CL↗

arXiv2023

Towards Reliable Dermatology Evaluation Benchmarks

Benchmark datasets for digital dermatology unwittingly contain inaccuracies that reduce trust in model performance estimates. We propose a resource-efficient data-cleaning protocol to identify issues that escaped previous curation. The protocol leverages an existing algorithmic cleaning strategy and is followed by a confirmation process terminated by an intuitive stopping criterion. Based on confirmation by multiple dermatologists, we remove irrelevant samples and near duplicates and estimate the percentage of label errors in six dermatology image datasets for model evaluation promoted by the International Skin Imaging Collaboration. Along with this paper, we publish revised file lists for each dataset which should be used for model evaluation. Our work paves the way for more trustworthy performance assessment in digital dermatology.

cs.CV↗

arXiv2023

Assessing Guest Nationality Composition from Hotel Reviews

Many hotels target guest acquisition efforts to specific markets in order to best anticipate individual preferences and needs of their guests. Likewise, such strategic positioning is a prerequisite for efficient marketing budget allocation. Official statistics report on the number of visitors from different countries, but no fine-grained information on the guest composition of individual businesses exists. There is, however, growing interest in such data from competitors, suppliers, researchers and the general public. We demonstrate how machine learning can be leveraged to extract references to guest nationalities from unstructured text reviews in order to dynamically assess and monitor the dynamics of guest composition of individual businesses. In particular, we show that a rather simple architecture of pre-trained embeddings and stacked LSTM layers provides a better performance-runtime tradeoff than more complex state-of-the-art language models.

cs.CL↗

arXiv2023

Robust T-Loss for Medical Image Segmentation

This paper presents a new robust loss function, the T-Loss, for medical image segmentation. The proposed loss is based on the negative log-likelihood of the Student-t distribution and can effectively handle outliers in the data by controlling its sensitivity with a single parameter. This parameter is updated during the backpropagation process, eliminating the need for additional computation or prior information about the level and spread of noisy labels. Our experiments show that the T-Loss outperforms traditional loss functions in terms of dice scores on two public medical datasets for skin lesion and lung segmentation. We also demonstrate the ability of T-Loss to handle different types of simulated label noise, resembling human error. Our results provide strong evidence that the T-Loss is a promising alternative for medical image segmentation where high levels of noise or outliers in the dataset are a typical phenomenon in practice. The project website can be found at https://robust-tloss.github.io

cs.CV↗

arXiv2023

Intrinsic Self-Supervision for Data Quality Audits

Benchmark datasets in computer vision often contain off-topic images, near duplicates, and label errors, leading to inaccurate estimates of model performance. In this paper, we revisit the task of data cleaning and formalize it as either a ranking problem, which significantly reduces human inspection effort, or a scoring problem, which allows for automated decisions based on score distributions. We find that a specific combination of context-aware self-supervised representation learning and distance-based indicators is effective in finding issues without annotation biases. This methodology, which we call SelfClean, surpasses state-of-the-art performance in detecting off-topic images, near duplicates, and label errors within widely-used image datasets, such as ImageNet-1k, Food-101N, and STL-10, both for synthetic issues and real contamination. We apply the detailed method to multiple image benchmarks, identify up to 16% of issues, and confirm an improvement in evaluation reliability upon cleaning. The official implementation can be found at: https://github.com/Digital-Dermatology/SelfClean.

cs.CV↗

arXiv2020

Cable Tree Wiring -- Benchmarking Solvers on a Real-World Scheduling Problem with a Variety of Precedence Constraints

Cable trees are used in industrial products to transmit energy and information between different product parts. To this date, they are mostly assembled by humans and only few automated manufacturing solutions exist using complex robotic machines. For these machines, the wiring plan has to be translated into a wiring sequence of cable plugging operations to be followed by the machine. In this paper, we study and formalize the problem of deriving the optimal wiring sequence for a given layout of a cable tree. We summarize our investigations to model this cable tree wiring Problem (CTW) as a traveling salesman problem with atomic, soft atomic, and disjunctive precedence constraints as well as tour-dependent edge costs such that it can be solved by state-of-the-art constraint programming (CP), Optimization Modulo Theories (OMT), and mixed-integer programming (MIP) solvers. It is further shown, how the CTW problem can be viewed as a soft version of the coupled tasks scheduling problem. We discuss various modeling variants for the problem, prove its NP-hardness, and empirically compare CP, OMT, and MIP solvers on a benchmark set of 278 instances. The complete benchmark set with all models and instance data is available on github and is accepted for inclusion in the MiniZinc challenge 2020.

cs.AI↗