1 citations · 2 across the 21 of their papers we have counts for
21 papers
Dense Structural Compression of Transformers via Gauge-Correct Channel Removal
Jed A. Duersch, Naïm Es-Sebbani, Nathanaël Haas +1
Inference energy per token drives the cost and carbon footprint of deployed transformers. It is dominated by dense matrix products that incur fused multiply-accumulate (FMA) operat…
Explanation Quality Assessment as Ranking with Listwise Rewards
Thomas Bailleux, Tanmoy Mukherjee, Emmanuel Lonca +2
We reformulate explanation quality assessment as a ranking problem rather than a generation problem. Instead of optimizing models to produce a single "best" explanation token-by-to…
Credal Concept Bottleneck Models for Epistemic-Aleatoric Uncertainty Decomposition
Tanmoy Mukherjee, Thomas Bailleux, Pierre Marquis +1
Concept Bottleneck Models (CBMs) predict through human-interpretable concepts, but they typically output point concept probabilities that conflate epistemic uncertainty (reducible…
Why Deep Jacobian Spectra Separate: Depth-Induced Scaling and Singular-Vector Alignment
Nathanaël Haas, François Gatine, Augustin M Cosse +1
Understanding why gradient-based training in deep networks exhibits strong implicit bias remains challenging, in part because tractable singular-value dynamics are typically availa…
Evaluating Robustness of Reasoning Models on Parameterized Logical Problems
Naïm Es-sebbani, Esteban Marquer, Yakoub Salhi +1
Logic provides a controlled testbed for evaluating LLM-based reasoners, yet standard SAT-style benchmarks often conflate surface difficulty (length, wording, clause order) with the…
Fourier Transformers for Latent Crystallographic Diffusion and Generative Modeling
Jed A. Duersch, Elohan Veillon, Astrid Klipfel +2
The discovery of new crystalline materials calls for generative models that handle periodic boundary conditions, crystallographic symmetries, and physical constraints, while scalin…
Structurally Separated Uncertainty in Supervised Latent Variable Models
Tanmoy Mukherjee, Marius Kloft, Pierre Marquis +1
Predictive uncertainty is commonly decomposed into epistemic and aleatoric components, but standard decompositions often produce strongly correlated estimates because both quantiti…
A Chemically Grounded Evaluation Framework for Generative Models in Materials Discovery
Elohan Veillon, Astrid Klipfel, Adlane Sayede +1
Generative models hold great promise for accelerating materials discovery, but their evaluation often overlooks the chemical validity and stability requirements crucial to real-wor…
Skim-Aware Contrastive Learning for Efficient Document Representation
Waheed Ahmed Abro, Zied Bouraoui
Although transformer-based models have shown strong performance in word- and sentence-level tasks, effectively representing long documents, especially in fields like law and medici…
MODE: Multi-Objective Adaptive Coreset Selection
Tanmoy Mukherjee, Pierre Marquis, Zied Bouraoui
We present Mode(Multi-Objective adaptive Data Efficiency), a framework that dynamically combines coreset selection strategies based on their evolving contribution to model performa…
Generalizing Analogical Inference from Boolean to Continuous Domains
Francisco Cunha, Yves Lepage, Miguel Couceiro +1
Analogical reasoning is a powerful inductive mechanism, widely used in human cognition and increasingly applied in artificial intelligence. Formal frameworks for analogical inferen…
Towards a Neurosymbolic Reasoning System Grounded in Schematic Representations
François Olivier, Zied Bouraoui
Despite significant progress in natural language understanding, Large Language Models (LLMs) remain error-prone when performing logical reasoning, often lacking the robust mental r…
Enhancing DR Classification with Swin Transformer and Shifted Window Attention
Meher Boulaabi, Takwa Ben Aïcha Gader, Afef Kacem Echi +1
Diabetic retinopathy (DR) is a leading cause of blindness worldwide, underscoring the importance of early detection for effective treatment. However, automated DR classification re…
Grounding Agent Reasoning in Image Schemas: A Neurosymbolic Approach to Embodied Cognition
François Olivier, Zied Bouraoui
Despite advances in embodied AI, agent reasoning systems still struggle to capture the fundamental conceptual structures that humans naturally use to understand and interact with t…
Modeling Multi-modal Cross-interaction for Multi-label Few-shot Image Classification Based on Local Feature Selection
Kun Yan, Zied Bouraoui, Fangyun Wei +4
The aim of multi-label few-shot image classification (ML-FSIC) is to assign semantic labels to images, in settings where only a small number of training examples are available for…
REFINE-LM: Mitigating Language Model Stereotypes via Reinforcement Learning
Rameez Qureshi, Naïm Es-Sebbani, Luis Galárraga +3
With the introduction of (large) language models, there has been significant concern about the unintended bias such models may inherit from their training data. A number of studies…
Ontology Completion with Natural Language Inference and Concept Embeddings: An Analysis
Na Li, Thomas Bailleux, Zied Bouraoui +1
We consider the problem of finding plausible knowledge that is missing from a given ontology, as a generalisation of the well-studied taxonomy expansion task. One line of work trea…
Modelling Commonsense Commonalities with Multi-Facet Concept Embeddings
Hanane Kteich, Na Li, Usashi Chatterjee +2
Concept embeddings offer a practical and efficient mechanism for injecting commonsense knowledge into downstream tasks. Their core purpose is often not to predict the commonsense p…
Vector Field Oriented Diffusion Model for Crystal Material Generation
Astrid Klipfel, Yaël Fregier, Adlane Sayede +1
Discovering crystal structures with specific chemical properties has become an increasingly important focus in material science. However, current models are limited in their abilit…
What do Deck Chairs and Sun Hats Have in Common? Uncovering Shared Properties in Large Concept Vocabularies
Amit Gajbhiye, Zied Bouraoui, Na Li +3
Concepts play a central role in many applications. This includes settings where concepts have to be modelled in the absence of sentence context. Previous work has therefore focused…
Inferring Prototypes for Multi-Label Few-Shot Image Classification with Word Vector Guided Attention
Kun Yan, Chenbin Zhang, Jun Hou +4
Multi-label few-shot image classification (ML-FSIC) is the task of assigning descriptive labels to previously unseen images, based on a small number of training examples. A key fea…