António Loison, Quentin Macé, Antoine Edy, Victor Xing, Tom Balough, Gabriel de Souza P. Moreira, Bo Liu, Manuel Faysse, Celine Hudelot, Gautier Viaud. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

In the trend of hybrid Artificial Intelligence techniques, Physics-Informed Machine Learning has seen a growing interest. It operates mainly by imposing data, learning, or architecture bias with simulation data, Partial Differential Equations, or equivariance and invariance properties. While it has shown great success on tasks involving one physical domain, such as fluid dynamics, existing methods are not adapted to tasks with complex multi-physical and multi-domain phenomena. In addition, it is mainly formulated as an end-to-end learning scheme. To address these challenges, we propose to leverage Bond Graphs, a multi-physics modeling approach, together with Graph Neural Networks. We propose a Neural Bond graph Encoder (NBgE) producing multi-physics-informed representations that can be fed into any task-specific model. It provides a unified way to integrate both data and architecture biases in deep learning. Our experiments on two challenging multi-domain physical systems – a Direct Current Motor and the Respiratory System – demonstrate the effectiveness of our approach on a multivariate time-series forecasting task.

This paper presents FactNET, a framework for human-centered granular fact-checking. Each claim to check is decomposed into sub claims that are fact-checked independently by using an LLM’s internal knowledge. The user interacts with the framework through an innovative interface, displaying the complex claim in the form of a graph, allowing the input of human knowledge on the topic, and evaluating the trust given in the generated evidence. A video demonstrating the system is available at TO_BE_PUBLISHED (in submission documents for review phase).

A limitation of modern document retrieval embedding methods is that they typically encode passages (chunks) from the same documents independently, often overlooking crucial contextual information from the rest of the document that could greatly improve individual chunk representations.In this work, we introduce ConTEB (Contextaware Text Embedding Benchmark), a benchmark designed to evaluate retrieval models on their ability to leverage document-wide context.Our results show that state-of-the-art embedding models struggle in retrieval scenarios where context is required.To address this limitation, we propose InSeNT (In-sequence Negative Training), a novel contrastive posttraining approach which combined with late chunking pooling enhances contextual representation learning while preserving computational efficiency.Our method significantly improves retrieval quality on ConTEB without sacrificing base model performance.We further find chunks embedded with our method are more robust to suboptimal chunking strategies and larger retrieval corpus sizes.We opensource all artifacts at https://github.com/ illuin-tech/contextual-embeddings.

Documents are visually rich structures that convey information through text, but also figures, page layouts, tables, or even fonts. Since modern retrieval systems mainly rely on the textual information they extract from document pages to index documents -often through lengthy and brittle processes-, they struggle to exploit key visual cues efficiently. This limits their capabilities in many practical document retrieval applications such as Retrieval Augmented Generation (RAG). To benchmark current systems on visually rich document retrieval, we introduce the Visual Document Retrieval Benchmark $\textit{ViDoRe}$, composed of various page-level retrieval tasks spanning multiple domains, languages, and practical settings. The inherent complexity and performance shortcomings of modern systems motivate a new concept; doing document retrieval by directly embedding the images of the document pages. We release $\textit{ColPali}$, a Vision Language Model trained to produce high-quality multi-vector embeddings from images of document pages. Combined with a late interaction matching mechanism, $\textit{ColPali}$ largely outperforms modern document retrieval pipelines while being drastically simpler, faster and end-to-end trainable. We release models, data, code and benchmarks under open licenses at https://hf.co/vidore.

Exemplar-free class-incremental learning (EFCIL) aims to classify streaming data without storing examples from the past. Recent EFCIL works suggest that (i) models pre-trained with large amounts of data should be used to initialize learning, (ii) self-supervised learned transformers generalize better than supervised convolutional models, (iii) adding generated data to the pre-training dataset can improve incremental accuracy. In this article, we question the above assertions by comprehensively evaluating various initial training strategies combined with four EFCIL algorithms using four large-scale datasets. Our results indicate that: (i) Pre-trained models are preferable when the domain of the incremental classification task is well represented in the pre-training datasets, but training with initial data remains useful when the domain shift is significant, (ii) supervised convolutional networks remain competitive, particularly when improving representation transferability using data augmentation or a projector, (iii) adding classes from an external dataset to train the initial model boosts performance when the initial set of classes is small but has a limited effect otherwise, (iv) additional classes generated with a diffusion model are not necessarily more useful than a well-chosen set of ImageNet classes to improve model transferability. We provide a nuanced analysis of these results and formulate recommendations to facilitate the practical adoption of EFCIL algorithms.

We address the problem of designing interpretable algorithms for image classification. Modern computer vision algorithms implement classification in two phases: feature extraction - the encoding - that relies on deep neural networks (DNN), followed by a task-oriented decision - the decoding - often also using a DNN. We propose to formulate this last phase as an argumentative DialoguE Between two agents relying on visual ATtributEs and Similarity to prototypes (DEBATES). DEBATES represents the combination of information provided by two encoders in a transparent and interpretable way. It relies on a dual process that combines similarity to prototypes and visual attributes, each extracted from an encoder. DEBATES makes explicit the agreements and conflicts between the two encoders managed by the two agents, reveals the causes of unintended behaviors, and helps identify potential corrective actions to improve performance. The approach is demonstrated on two problems of fine-grained image classification.

Class-Incremental Learning (CIL) aims to build classification models from data streams. At each step of the CIL process, new classes must be integrated into the model. Due to catastrophic forgetting, CIL is particularly challenging when examples from past classes cannot be stored, the case on which we focus here. To date, most approaches are based exclusively on the target dataset of the CIL process. However, the use of models pre-trained in a self-supervised way on large amounts of data has recently gained momentum. The initial model of the CIL process may only use the first batch of the target dataset, or also use pre-trained weights obtained on an auxiliary dataset. The choice between these two initial learning strategies can significantly influence the performance of the incremental learning model, but has not yet been studied in depth. Performance is also influenced by the choice of the CIL algorithm, the neural architecture, the nature of the target task, the distribution of classes in the stream and the number of examples available for learning. We conduct a comprehensive experimental study to assess the roles of these factors. We present a statistical analysis framework that quantifies the relative contribution of each factor to incremental performance. Our main finding is that the initial training strategy is the dominant factor influencing the average incremental accuracy, but that the choice of CIL algorithm is more important in preventing forgetting. Based on this analysis, we propose practical recommendations for choosing the right initial training strategy for a given incremental learning use case. These recommendations are intended to facilitate the practical deployment of incremental learning.

In order to foster inclusive education, automatic systems that can adapt textbooks to make them accessible to children with Developmental Coordination Disorder (DCD) are necessary. In this context, we propose a task to classify exercises according to their DCD adaptation type. We introduce a challenging exercise dataset extracted from French textbooks, with two major difficulties: limited and unbalanced, noisy data. To set a baseline on the dataset, we use state-of-the-art models combined through early and late fusion techniques to take advantage of text and vision/layout modalities. Our approach achieves an overall accuracy of 0.802. However, the experiments show the difficulty of the task, especially for minority classes, where the accuracy drops to 0.583.

We tackle the Few-Shot Open-Set Recognition (FSOSR) problem, i.e. classifying instances among a set of classes for which we only have a few labeled samples, while simultaneously detecting instances that do not belong to any known class. We explore the popular transductive setting, which leverages the unlabelled query instances at inference. Motivated by the observation that existing transductive methods perform poorly in open-set scenarios, we propose a generalization of the maximum likelihood principle, in which latent scores down-weighing the influence of potential outliers are introduced alongside the usual parametric model. Our formulation embeds supervision constraints from the support set and additional penalties discouraging overconfident predictions on the query set. We proceed with a block-coordinate descent, with the latent scores and parametric model co-optimized alternately, thereby benefiting from each other. We call our resulting formulation Open-Set Likelihood Optimization (OSLO). OSLO is interpretable and fully modular; it can be applied on top of any pre-trained model seamlessly. Through extensive experiments, we show that our method surpasses existing inductive and transductive methods on both aspects of open-set recognition, namely inlier classification and outlier detection. Code is available at https://github.com/ebennequin/few-shot-open-set.

Instruction Fine-Tuning (IFT) is a powerful paradigm that strengthens the zero-shot capabilities of Large Language Models (LLMs), but in doing so induces new evaluation metric requirements.We show LLM-based metrics to be well adapted to these requirements, and leverage them to conduct an investigation of taskspecialization strategies, quantifying the tradeoffs that emerge in practical industrial settings.Our findings offer practitioners actionable insights for real-world IFT model deployment.

Recent class-incremental learning methods combine deep neural architectures and learning algorithms to handle streaming data under memory and computational constraints. The performance of existing methods varies depending on the characteristics of the incremental process. To date, there is no other approach than to test all pairs of learning algorithms and neural architectures on the training data available at the start of the learning process to select a suited algorithm-architecture combination. To tackle this problem, in this article, we introduce AdvisIL, a method which takes as input the main characteristics of the incremental process (memory budget for the deep model, initial number of classes, size of incremental steps) and recommends an adapted pair of learning algorithm and neural architecture. The recommendation is based on a similarity between the user-provided settings and a large set of pre-computed experiments. AdvisIL makes class-incremental learning easier, since users do not need to run cumbersome experiments to design their system. We evaluate our method on four datasets under six incremental settings and three deep model sizes. We compare six algorithms and three deep neural architectures. Results show that AdvisIL has better overall performance than any of the individual combinations of a learning algorithm and a neural architecture. AdvisIL’s code is available at https://github.com/EvaJF/AdvisIL.

With the increase in use of machine learning classifiers in several fields, providing human- understandable explanation of their outputs has become an imperative. It is essential to generate trust for day-to-day tasks, especially in the sensible domains as medical imaging. Although many works have addressed this problem by generating visual explanation maps, they often provide noisy and inaccurate results forcing heuristic regularization unrelated to the classifier in question. In this paper, we propose a general perspective of the visual explanation problem overcoming these limitations. We show that visual explanation can be produced as the difference between two generated images obtained via two specific conditional generative models. Both generative models are trained using the classifier to explain and a database to enforce the following properties: (i) All images generated by the first generator are classified similarly to the input image, whereas the second generator’s outputs are classified oppositely. (ii) All generated images belong to the distribution of real images. (iii) The distances between the input image and the corresponding generated images are minimal so that the difference between the generated elements only reveals relevant information for the studied classifier. Using symmetrical and cyclic constraints, we present two different approximations and implementations of the general formulation. Experimentally, we demonstrate significant improvements with respect to the state-of-the-art on three different public data sets. In particular, the localization of regions influencing the classifier is consistent with human annotations.

Recently, due to the ubiquity and supremacy of E-recruitment platforms, job recommender systems have been largely studied. In this paper, we tackle the next job application problem, which has many practical applications. In particular, we propose to leverage next-item recommendation approaches to consider better the job seeker’s career preference to discover the next relevant job postings (referred to jobs for short) they might apply for. Our proposed model, named Personalized-Attention Next-Application Prediction (PANAP), is composed of three modules. The first module learns job representations from textual content and metadata attributes in an unsupervised way. The second module learns job seeker representations. It includes a personalized-attention mechanism that can adapt the importance of each job in the learned career preference representation to the specific job seeker’s profile. The attention mechanism also brings some interpretability to learned representations. Then, the third module models the Next-Application Prediction task as a top-K search process based on the similarity of representations. In addition, the geographic location is an essential factor that affects the preferences of job seekers in the recruitment domain. Therefore, we explore the influence of geographic location on the model performance from the perspective of negative sampling strategies. Experiments on the public CareerBuilder12 dataset show the interest in our approach.

Domain Invariant Representations (IR) has improved drastically the transferability of representations from a labelled source domain to a new and unlabelled target domain. Unsupervised Domain Adaptation (UDA) in presence of label shift remains an open problem. To this purpose, we present a bound of the target risk which incorporates both weights and invariant representations. Our theoretical analysis highlights the role of inductive bias in aligning distributions across domains. We illustrate it on standard benchmarks by proposing a new learning procedure for UDA. We observed empirically that weak inductive bias makes adaptation robust to label shift. The elaboration of stronger inductive bias is a promising direction for new UDA algorithms.

In this paper, we present a novel cross-consistency based semi-supervised approach for semantic segmentation. Consistency training has proven to be a powerful semi-supervised learning framework for leveraging unlabeled data under the cluster assumption, in which the decision boundary should lie in low-density regions. In this work, we first observe that for semantic segmentation, the low-density regions are more apparent within the hidden representations than within the inputs. We thus propose cross-consistency training, where an invariance of the predictions is enforced over different perturbations applied to the outputs of the encoder. Concretely, a shared encoder and a main decoder are trained in a supervised manner using the available labeled examples. To leverage the unlabeled examples, we enforce a consistency between the main decoder predictions and those of the auxiliary decoders, taking as inputs different perturbed versions of the encoder's output, and consequently, improving the encoder's representations. The proposed method is simple and can easily be extended to use additional training signal, such as image-level labels or pixel-level labels across different domains. We perform an ablation study to tease apart the effectiveness of each component, and conduct extensive experiments to demonstrate that our method achieves state-of-the-art results in several datasets.

Active learning aims to optimize the dataset annotation process when resources are constrained. Most existing methods are designed for balanced datasets. Their practical applicability is limited by the fact that a majority of real-life datasets are actually imbalanced. Here, we introduce a new active learning method which is designed for imbalanced datasets. It favors samples likely to be in minority classes so as to reduce the imbalance of the labeled subset and create a better representation for these classes. We also compare two training schemes for active learning: (1) the one commonly deployed in deep active learning using model fine tuning for each iteration and (2) a scheme which is inspired by transfer learning and exploits generic pre-trained models and train shallow classifiers for each iteration. Evaluation is run with three imbalanced image datasets. Results show that the proposed active learning method outperforms competitive baselines. Equally interesting, they also indicate that the transfer learning training scheme outperforms model fine tuning if features are transferable from the generic dataset to the unlabeled one. This last result is surprising and should encourage the community to explore the design of deep active learning methods.

Explaining decisions of black-box classifiers is paramount in sensitive domains such as medical imaging since clinicians confidence is necessary for adoption. Various explanation approaches have been proposed, among which perturbation based approaches are very promising. Within this class of methods, we leverage a learning framework to produce our visual explanations method. From a given classifier, we train two generators to produce from an input image the so called similar and adversarial images. The similar image shall be classified as the input image whereas the adversarial shall not. Visual explanation is built as the difference between these two generated images. Using metrics from the literature, our method outperforms state-of-the-art approaches. The proposed approach is model-agnostic and has a low computation burden at prediction time. Thus, it is adapted for real-time systems. Finally, we show that random geometric augmentations applied to the original image play a regularization role that improves several previously proposed explanation methods. We validate our approach on a large chest X-ray database.

Among the wide variety of image generative models, two models stand out: Variational Auto Encoders (VAE) and Generative Adversarial Networks (GAN). GANs can produce realistic images, but they suffer from mode collapse and do not provide simple ways to get the latent representation of an image. On the other hand, VAEs do not have these problems, but they often generate images less realistic than GANs. In this article, we explain that this lack of realism is partially due to a common underestimation of the natural image manifold dimensionality. To solve this issue we introduce a new framework that combines VAE and GAN in a novel and complementary way to produce an auto-encoding model that keeps VAEs properties while generating images of GAN-quality. We evaluate our approach both qualitatively and quantitatively on five image datasets.

We present Event2018, a corpus annotated for event detection tasks, consisting of 38 million tweets in French (retweets excluded) including more than 130,000 tweets manually annotated by three annotators as related or unrelated to a given event. The 243 events were selected both from press articles and from subjects trending on Twitter during the annotation period (July to August 2018). In total, more than 95,000 tweets were annotated as related to one of the selected events. We also provide the titles and URLs of 15,500 news articles automatically detected as related to these events. In addition to this corpus, we detail the results of our event detection experiments on both this dataset and another publicly available dataset of tweets in English. We ran extensive tests with different types of text embeddings and a standard Topic Detection and Tracking algorithm, and detail our evaluation method. We show that tf-idf vectors allow the best performance for this task on both corpora. These results are intended to serve as a baseline for researchers wishing to test their own event detection systems on our corpus.

Active learning increases the effectiveness of labeling when only subsets of unlabeled datasets can be processed manually. To our knowledge, existing algorithms are designed under the assumption that datasets are balanced. However, many real-life datasets are actually imbalanced and we propose two adaptations of active learning to tackle imbalance. First, we modify acquisition functions to select samples by taking advantage of a deep model pretrained on a source domain. Second, we introduce a balancing step in the acquisition process to reduce the imbalance of the labeled subset. Evaluation is done with four imbalanced datasets using existing active learning methods and their modifications introduced here. Results show that our adaptations are useful as long as knowledge from the source domain is transferable to target domains.

A representation is supposed universal if it encodes any element of the visual world (e.g., objects, scenes) in any configuration (e.g., scale, context). While not expecting pure universal representations, the goal in the literature is to improve the universality level, starting from a representation with a certain level. To improve that universality level, one can diversify the source-task, but it requires many additive annotated data that is costly in terms of manual work and possible expertise. We formalize such a diversification process then propose two methods to improve the universality of CNN representations that limit the need for additive annotated data. The first relies on human categorization knowledge and the second on re-training using fine-tuning. We propose a new aggregating metric to evaluate the universality in a transfer-learning scheme, that addresses more aspects than previous works. Based on it, we show the interest of our methods on 10 target-problems, relating to classification on a variety of visual domains.

The goal of explainable artificial intelligence is to solve problems in a way that humans can understand how it does it. However, few approaches have been proposed so far and some of them lay more emphasis on interpretability than on explainability. In this paper, we propose an approach that is based on learning fuzzy relations and fuzzy properties. We extract frequent relations from a dataset to generate an explained decision. Our approach can deal with different problems, such as classification or annotation. A model was built to perform explained classification on a toy dataset that we generated. It managed to correctly classify examples while providing convincing explanations. A few areas for improvement have been spotted, such as the need to filter relations and properties before or while learning them in order to avoid useless computations.

In a transfer-learning scheme, the intermediate layers of a pre-trained CNN are employed as universal image representation to tackle many visual classification problems. The current trend to generate such representation is to learn a CNN on a large set of images labeled among the most specific categories. Such processes ignore potential relations between categories, as well as the categorical-levels used by humans to classify. In this paper, we propose Multi Categorical-Level Networks (MuCaLe-Net) that include human-categorization knowledge into the CNN learning process. A MuCaLe-Net separates generic categories from each other while it independently distinguishes specific ones. It thereby generates different features in the intermediate layers that are complementary when combined together. Advantageously, our method does not require additive data nor annotation to train the network. The extensive experiments over four publicly available benchmarks of image classification exhibit state-of-the-art performances.

We consider the problem of image classification with semantic features that are built from a set of base classifier outputs, each reflecting visual concepts. However, existing approaches consider visual concepts independently from each other whereas they are often linked together. When those relations are considered, existing models strongly rely on image low-level features, yielding in irrelevant relations when the low-level representation fails. On the contrary, the approach we propose, uses existing human knowledge, the application context itself and the human categorization mechanism to reflect complex relations between concepts. By nesting this human knowledge and the application context in the concept detection and selection processes, our final semantic feature captures the most useful information for an effective categorization. Thus, it enables to give good representation, even if some important concepts failed to be recognized. Experimental validation is conducted on three publicly available benchmarks of multi-class object classification and leads to results that outperforms comparable approaches.

In this paper we extend some previously established links between the derivation operators used in formal concept analysis and some mathematical morphology operators to fuzzy concept analysis. We also propose to use mathematical morphology to navigate in a fuzzy concept lattice and perform operations on it. Links with other lattice-based for malisms such as rough sets and F-transforms are also established. This paper proposes a discussion and new results on such links and their potential interest.

We propose an ontology alignment framework with two core features: the use of background knowledge and the ability to handle vagueness in the matching process and the resulting concept alignments. The procedure is based on the use of a generic reference vocabulary, which is used for fuzzifying the ontologies to be matched. The choice of this vocabulary is problem-dependent in general, although Wikipedia represents a general-purpose source of knowledge that can be used in many cases, and even allows cross language matchings. In the first step of our approach, each domain concept is represented as a fuzzy set of reference concepts. In the next step, the fuzzified domain concepts are matched to one another, resulting in fuzzy descriptions of the matches of the original concepts. Based on these concept matches, we propose an algorithm that produces a merged fuzzy ontology that captures what is common to the source ontologies. The paper describes experiments in the domain of multimedia by using ontologies containing tagged images, as well as an evaluation of the approach in an information retrieval setting. The undertaken fuzzy approach has been compared to a classical crisp alignment by the help of a ground truth that was created based on human judgment.

In this paper, we propose an original way of enriching description logics with abduction reasoning services. Under the aegis of set and lattice theories, we put together ingredients from mathematical morphology, description logics, and formal concept analysis. We propose computing the best explanations of an observation through algebraic erosion over the concept lattice of a background theory that is efficiently constructed using tools from formal concept analysis. We show that the defined operators are sound and complete and satisfy important rationality postulates of abductive reasoning. As a typical illustration, we consider a scene understanding problem. In fact, scene understanding can benefit from prior structural knowledge represented as an ontology and the reasoning tools of description logics. We formulate model based scene understanding as an abductive reasoning process. A scene is viewed as an observation and the interpretation is defined as the best explanation, considering the terminological knowledge part of a description logic about the scene context. This explanation is obtained from morphological operators applied on the corresponding concept lattice.

We address the problem of tag completion for automatic image annotation. Our method consists in two main steps: creating a list of "candidate tags" from the visual neighbors of the untagged image then using them as pieces of evidence to be combined to provide the final list of predicted tags. Both steps introduce a scheme to tackle with imprecision and uncertainty. First, a bag-of-words (BOW) signature is generated for each neighbor using local soft coding. Second, a sum-pooling operation across the BOW of the k nearest neighbors provides the list of "candidate tags". Finally, we use neighbors as pieces of evidence to be combined according to the Dempster's rule to predict the more relevant tags. The method is evaluated in the context of image classification and that of tag suggestion. The database used for visual neighbors search contains 1.2 million images extracted from Flickr. Classification is evaluated on the well known Pascal VOC 2007 and MIR Flickr datasets, on which we obtain similar or better results than the state-of-the-art. For tag suggestion, we manually annotated 241 queries. As well, we obtain competitive results on this task.

Semantic hierarchies have been introduced recently to improve image annotation. They was used as a framework for hierarchical image classification, and thus to improve classifiers accuracy and reduce the complexity of managing large scale data. In this paper, we investigate the contribution of semantic hierarchies for hierarchical image classification. We propose first a new method based on the hierarchy structure to train efficiently hierarchical classifiers. Our method, named One-Versus-Opposite-Nodes, allows decomposing the problem in several independent tasks and therefore scales well with large database. We also propose two methods for computing a hierarchical decision function that serves to annotate new image samples. The former is performed by a top-down classifiers voting, while the second is based on a bottom-up score fusion. The experiments on Pascal VOC'2010 dataset showed that our methods improve well the image annotation results.

The automatic attribution of semantic labels to unlabeled or weakly labeled images has received considerable attention but, given the complexity of the problem, remains a hard research topic. Here we propose a unified classification framework which mixes textual and visual information in a seamless manner. Unlike most recent previous works, computer vision techniques are used as inspiration to process textual information. To do so, we consider two types of complementary tag similarities, respectively computed from a conceptual hierarchy and from data collected from a photo sharing platform. Visual content is processed using recent techniques for bag-of visual-words feature generation. A central contribution of our work is to infer the coding step of the general bag-of-word framework with such similarities and to aggregate these tag-codes by max-pooling to obtain a single representative vector (signature). Final image annotations are obtained via late fusion, where the three modalities (two text-based and one visual-based) are merged during the classification step. Experimental results on the Pascal VOC 2007 and MIR Flickr datasets show an improvement over the state-of-the-art methods, while significantly decreasing the computational complexity of the learning system.

Abstract In this paper, we introduce a new formalism for mesh geometry prediction. We derive a class of smooth linear predictors from a simple approach based on the Taylor expansion of the mesh geometry function. We use this method as a generic way to compute weights for various linear predictors used for mesh compression and compare them with those of existing methods. We show that our scheme is actually equivalent to the Modified Butterfly subdivision scheme used for wavelet mesh compression. We also build new efficient predictors that can be used for connectivity‐driven compression in place of other schemes like Average/Dual Parallelogram Prediction and High Degree Polygon Prediction. The new predictors use the same neighbourhood, but do not make any assumption on mesh anisotropy. In the case of Average Parallelogram Prediction, our new weights improve compression rates from 3% to 18% on our test meshes. For Dual Parallelogram Prediction, our weights are equivalent to those of the previous Freelence approach, that outperforms traditional schemes by 16% on average. Our method effectively shows that these weights are optimal for the class of smooth meshes. Modifying existing schemes to make use of our method is free because only the prediction weights have to be modified in the code.

Bipolarity is an important feature of spatial information, involved in the expression of preferences and constraints about spatial positioning or in pairs of opposite spatial relations such as left and right. Another important feature is imprecision which has to be taken into account to model vagueness, inherent to many spatial relations (as for instance vague expressions such as close to, to the right of), and to gain in robustness in the representations. In previous works, we have shown that fuzzy sets and fuzzy mathematical morphology are appropriate frameworks, on the one hand, to represent bipolarity and imprecision of spatial relations and, on the other hand, to combine qualitative and quantitative reasoning in description logics extended with fuzzy concrete domains. The purpose of this paper is to integrate the bipolarity feature in the latter logical framework based on bipolar and fuzzy mathematical morphology and description logics with fuzzy concrete domains. Two important issues are addressed in this paper: the modeling of the bipolarity of spatial relations at the terminological level and the integration of bipolar notions in fuzzy description logics. At last, we illustrate the potential of the proposed formalism for spatial reasoning on a simple example in brain imaging.

The linguistic description, i.e. semantic annotation of images can benefit from representations of useful concepts and the links between them as ontologies. Recently, several multimedia ontologies have been proposed in the literature as suitable knowledge models to bridge the well known semantic gap between low level features of image content and its high level conceptual meaning. Nevertheless, these multimedia ontologies are often dedicated to (or initially built for) particular needs or a particular application. Ontology matching, defined as the process of relating different heterogeneous models, could be a suitable approach to solve several interoperability issues that coexist in semantic image annotation and retrieval. In this paper, we propose an original and generic instance-based ontology matching approach and a methodology to extract a minimal ontology defined as the common reference between different heterogeneous ontologies. Then, this approach is applied to two different semantic image retrieval issues: the bridging of the semantic gap by the matching of a multimedia ontology with a common-sense knowledge ontology and the matching of different multimedia ontologies to extract a common reference knowledge model dedicated to several multimedia applications.

Abstract This paper presents a novel algorithm for hierarchical random accessible mesh decompression. Our approach progressively decompresses the requested parts of a mesh without decoding less interesting parts. Previous approaches divided a mesh into independently compressed charts and a base coarse mesh. We propose a novel hierarchical representation of the mesh. We build this representation by using a boundary‐based approach to recursively split the mesh in two parts, under the constraint that any of the two resulting submeshes should be reconstructible independently. In addition to this decomposition technique, we introduce the concepts of opposite vertex and context dependant numbering. This enables us to achieve seemingly better compression ratios than previous work on quad and higher degree polygonal meshes. Our coder uses about 3 bits per polygon for connectivity and 14 bits per vertex for geometry using 12 bits quantification.

This work presents a new object categorization method and shows how it can be used for image retrieval. Our approach involves machine learning and knowledge representation techniques. A major element of our approach is a visual concept ontology composed of several types of concepts (spatial concepts and relations, color concepts and texture concepts). Visual concepts contained in this ontology can be seen as an intermediate layer between domain knowledge and image processing procedures. Our approach is composed of three phases: (1) a knowledge acquisition phase, (2) a learning phase and (3) a categorization phase. This work is mainly focused on phases (2) and (3). A major issue is the symbol grounding problem which consists of linking meaningfully symbols to sensory information. We propose a solution to this difficult issue by showing how learning techniques can map numerical features to visual concepts.

This paper presents a generic cognitive vision platform for the automatic recognition of natural complex objects. The recognition consists of three steps : image processing for numerical object description, mapping of numerical data into symbolic data and semantic interpretation for object recognition. The focus of this paper is the distributed platform architecture composed of three highly specialized knowledge based systems (KBS). The first KBS is dedicated to semantic interpretation. The second one has to deal with the anchoring of symbolic data into image data. The last KBS is dedicated to intelligent image processing. After a brief overview of the natural object recognition problem, this paper describes the three subcomponents of the platform.